02831 PRF29210.1177/0267659113502831PerfusionTagaya et al.

Original Paper

Perfusion
2014, Vol. 29(2) 117­–123
Prospects for using a hemoconcentrator © The Author(s) 2013
Reprints and permissions:
as an alternative hemodialysis method in sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0267659113502831
cardiopulmonary bypass surgeries prf.sagepub.com

M Tagaya,1,2,3 M Matsuda,4 M Yakehiro3 and H Izutani5

Abstract
Objective: Cardioplegic solutions often cause high blood concentrations of potassium. The conventional hemoconcen-
tration circuit was improved to correct electrolyte imbalances through a method involving dilutional ultrafiltration
(DUF) and an alternative hemodialysis (ALTHD) method. This study aimed to determine the effectiveness of this
ALTHD method.
Methods: Bovine blood was used, in conjunction with a hemoconcentrator, in an experimental hemodialysis (HD) circuit
to evaluate an ALTHD method. The effectiveness of the method was determined by electrolyte and hematocrit meas-
urements following the procedure.
Results: The ALTHD method corrected electrolyte levels as effectively as DUF and was less affected by dilution than DUF.
Conclusion: The ALTHD method may provide faster electrolyte adjustments than DUF because its efficiency depends
on both the blood and dialysate flow rates. In addition, the ALTHD method is expected to provide increased efficiency.
Thus, our DUF/ALTHD circuit-switching method may be clinically useful when rapid electrolyte correction is required.

Keywords
cardiopulmonary bypass; hyperkalemia; hemodialysis; hemoconcentrator; dilutional ultrafiltration

Introduction
High-potassium cardioplegic solutions are used to arrest
cardiac action during open heart surgery, enabling favor-
able myocardial protection.1-4 However, high-potassium
cardioplegic solutions produce high serum potassium
levels, often resulting in significant challenges associated
with the maintenance of appropriate serum potassium
levels during extracorporeal circulation (ECC).5,6 In
Japan, control of serum potassium concentrations dur-
ing ECC is generally performed via a process involving
repeated blood dilution and concentration, called dilu-
generally for non-dialysis patients. Part of the reason
that HD is not more commonly used, in spite of its ben-
efits, may be due to the requirement for additional
hemodialysis, monitoring and water-treatment equip-
ment;8 the method also requires a longer, more compli-
1Clinical Engineer Center, Ehime University Hospital, Ehime, Japan
2Medical Equipment Management Office, National Hospital Organization
Kure Medical Center, Hiroshima, Japan
3Division of Physiology, Hiroshima International University, Hiroshima,

tional ultrafiltration (DUF), using a hemoconcentrator Japan

(Figure 1a). 4Department of Cardiology, National Hospital Organization Kure

The speed of DUF cannot be actively controlled as it is Medical Center, Hiroshima, Japan
5Department of Cardiovascular and Thoracic Surgery, Ehime University
dependent on the blood flow through the hemoconcen-
Graduate School of Medicine, Ehime, Japan
trator, thus, making the rapid adjustment of serum
potassium concentrations impossible. On the other Corresponding author:
hand, hemodialysis (HD) facilitates the rapid control of Masashi Tagaya
potassium levels by allowing alterations of the flow rate Clinical Engineer Center
Ehime University Hospital
of the dialysis solution, independent of the blood flow.7 454 Shitsukawa, Toon, Ehime 791-0295
Therefore, intraoperative HD is often used dur- Japan.
ing cardiac surgeries for dialysis-patients,8,9 but not Email: tagayam@kure-nh.go.jp
118 Perfusion 29(2)

Figure 1.  Methods for correcting electrolyte imbalances. (a) Traditional hemoconcentration circuit. The circuit is commonly
incorporated into the extracorporeal circulation system during cardiopulmonary bypass surgery. The process of repeated
concentration, after dilution, is called dilutional ultrafiltration. (b) A common hemodialysis (HD) circuit. In this method, the dialysis
solution is regulated by monitoring the equipment. HD requires more equipment for providing reverse osmosis of water and
providing undiluted dialysis solution as compared to the traditional hemoconcentration circuit. (c) The dilutional ultrafiltration/
hemodialysis switching circuit. The circuit switches between the 2 processes, based on the clamp position.

cated set-up (Figure 1b). All of these factors contribute to
increasing the expense associated with this procedure.
One would expect that the hemoconcentrator would
have the potential to produce a diffusional effect as effec-
tively as a hemodialyzer because both devices are simi-
larly constructed, differing only in the size of the pores in
the hollow fibers. However, hemoconcentrators are
rarely replaced by hemodialyzers during actual surgeries
because they do not facilitate blood concentration at the
same time as the HD. Therefore, reports on intraopera-
tive HD, using a hemoconcentrator as an alternate to a
hemodialyzer, have not been published.
In the present study, we designed a new method that
may resolve the challenge of performing intraoperative
HD by using a hemoconcentrator instead of a hemodia-
lyzer; the conventional hemoconcentration circuit was
simply improved by adding a switching junction, with-
out any additional equipment. This method is referred to
as a “circuit-switching DUF/HD process” or ALTHD
(Figure 1c). The primary advantage of this method is the
ability to switch between hemoconcentration, using the
DUF circuit, and an HD circuit, using a hemocon-
centrator instead of a hemodialyzer, depending on
the patient requirement. We investigated whether a
Tagaya et al. 119

Figure 2.  Experimental circuits. (a) Experimental circuit used for diffusion experiments. The circuit was shifted to a single-path
circuit by clamping the recirculation line after 5 min of recirculation. (b) Experimental dilutional ultrafiltration (DUF) circuit for
comparing the efficiency of DUF and hemodialysis. Saline was added to the reservoir and the diluted blood was filtered through the
hemoconcentrator. (c) Experimental hemodialysis (HD) circuit for comparing the efficiency of dilutional ultrafiltration and HD. Saline
was sent to the outlet of the hemoconcentrator for diffusion.

hemoconcentrator is able to produce a diffusional effect,
similar to that produced by a hemodialyzer, and evalu-
ated this ALTHD technique by comparing it with the tra-
ditional DUF method in an in vitro system.
Method
Diffusion with the hemoconcentrator
Variances in blood electrolytes were monitored by sam-
pling at the inlet and outlet of the hemoconcentrator
while maintaining blood flow within the hollow fibers
and saline flow outside of the hollow fibers. This was
accomplished using the following methodology.
Bovine blood (5 L), calibrated to a hematocrit (Hct) of
25%, a sodium (Na+) concentration of 144 mEq/L, a
potassium (K+) level of 8.0 mEq/L and a base excess (BE)
of -15 mEq/L, was maintained along with a temperature
of 37°C, with continuous stirring. The blood was recir-
culated through a hemoconcentrator (Aqua Stream,
JMS, Tokyo, Japan) at 300 mL/min in an experimental
circuit (Figure 2a) equipped with roller pumps and 3
sampling ports. After 5 min of saline recirculation to
prime the circuit, the recirculation line was clamped and
switched to a single-path system to eliminate any errors
caused by priming. Blood flow was maintained inside of
the hollow fibers (within the hemoconcentrator) at 300
mL/min and saline flow was maintained, outside of the
hollow fibers, at 50, 100 or 150 mL/min, thus, providing
a gradient for electrolyte transfer between the blood and
saline. After a 3-min flow period at each flow rate, the
blood was sampled at both the inlet and outlet of the
hemoconcentrator and the saline solution was sampled
at the outlet of the hemoconcentrator to determine elec-
trolyte concentrations. The levels of Na+, K+, BE and the
Hct were measured using an iSTAT1 analyzer (Abbott
Point Care, Princeton, NJ, USA).
Comparison of DUF and ALTHD
This experiment compared the variance in blood elec-
trolytes through identical DUF and ALTHD processes.
120 Perfusion 29(2)

Table 1.  Hemoconcentrator diffusion results.

Flow rate (Saline) Sampling Blood analysis 

  Na+ K+ BE Hct

  meq/L meq/L meq/L %

50 mL/min Blood in 143 8.0 –16 23
  Blood out 147 6.3 –19 24
  Saline out 146 7.8 –15 0
  Δ Blood in-out 4 1.7 3 1
100 mL/min Blood in 143 8.0 –16 23
  Blood out 151 4.6 –21 25
  Saline out 147 7.4 –15 0
  Δ Blood in-out 8 3.4 5 2
150 mL/min Blood in 143 8.0 –16 22
  Blood out 153 3.5 –22 26
  Saline out 149 6.5 –17 0
  Δ Blood in-out 10 4.5 6 4
Δ Blood in-out refers to the change in hematocrit and each measured electrolyte.

Each process was allowed to run for the same length of

time and involved a similar consumption of saline; this
allowed a comparison of the efficiency curves for both
methods.
Bovine blood (30 L) was calibrated to the same levels
as described previously and divided into 6 aliquots. The
DUF experimental circuit consisted of roller pumps, a
hemoconcentrator and a sampling port, similar to that of
the ALTHD circuit (Figures 2b and 2c). Each 5-L blood
volume was continuously stirred using a magnetic stirrer
and 3 aliquots were recirculated (300 mL/min) through
the DUF circuit and another 3 aliquots were recirculated
through the ALTHD circuit. After a 5-min recirculation
period (to eliminate any errors caused by the saline
priming), the blood was sampled to acquire the baseline
data. The DUF circuits were then recirculated at 300 mL/
min until each 3-L volume of saline (consecutively
pumped at 50, 100 or 150 mL/min) was expended. Figure 3.  Efficiency curve comparing saline flow rate and
Similarly, the ALTHD circuits were recirculated until diffusion efficiency (ΔNa+, change in sodium ion concentration;
each 3-L volume of saline (also consecutively pumped at ΔK+, change in potassium ion concentration; ΔBE, change in base
excess; ΔHct, change in hematocrit).
50, 100 or 150 mL/min) was expended. Each circuit was
sampled at the inlet of the hemoconcentrator to measure
the electrolyte concentrations at 3 time points (after
expending 1, 2 and 3 L of saline). The same parameters, the differences between inlet and outlet samples. Figure
as described above, were measured at each time point 3 shows the correlation of the saline flow rate with the
using the iSTAT1 analyzer. blood electrolyte variances. Based on the similarity of the
measurements obtained at different flow rates, ALTHD
with saline as the dialysis solution appears to be a suit-
Results able replacement for a hemoconcentrator.
Diffusion with the hemoconcentrator
Comparison of DUF and ALTHD
Table 1 shows the electrolyte levels and hematocrit of
blood sampled at the blood inlet and outlet ports as well Table 2 shows the variance in electrolyte and hematocrit
as the analysis results of saline from the outlet port. The levels, according to the DUF and ALTHD flow parame-
changes in concentrations reflect the absolute values of ters. Figure 4 indicates that there was little difference in
Tagaya et al. 121

Table 2.  Impact of dilutional ultrafiltration (DUF) and hemodialysis (HD) on the levels of electrolytes in an in vitro system.

Method Saline Blood analysis 

  flow rate volume Na+ K+ BE Hct
  mL/min L meq/L meq/L meq/L %
DUF  50 0 142 8.1 –15 23
  1 147 6.0 –19 24
  2 151 4.4 –22 25
  3 154 3.1 –24 27
  100 0 143 7.9 –15 22
  1 148 6.1 –19 24
  2 151 4.5 –22 24
  3 154 3.3 –24 26
  150 0 142 8.0 –15 22
  1 148 6.2 –19 22
  2 150 4.6 –21 21
  3 153 3.9 –23 22
HD  50 0 142 8.0 –16 23
  1 146 6.2 –19 25
  2 150 4.6 –21 26
  3 152 3.4 –24 29
  100 0 143 8.0 –16 22
  1 147 6.2 –19 24
  2 151 4.6 –21 25
  3 153 3.4 –23 27
  150 0 143 8.0 –16 23
  1 146 6.4 –18 24
  2 150 5.0 –21 25
  3 151 4.1 –22 25

0 L volume indicates baseline data; Hct, hematocrit; BE, base excess.

the efficiency of the electrolyte adjustments between
DUF and ALTHD. The hematocrit of the blood only fell
to a lower level during DUF than during ALTHD at a
saline flow rate of 150 mL/min.
Discussion
The results of this experiment demonstrated that the
ALTHD method has the ability to correct electrolyte lev-
els to a similar extent as DUF, at the same rates of blood
flow, while exhausting the saline solution over the same
period. In other words, the ALTHD method does not
depend on blood flow and may have the potential to cor-
rect electrolyte balances faster than DUF. Miyahara et al.8
reported that HD is more efficient than DUF for correct-
ing blood urea nitrogen, creatinine and electrolyte levels.
Thus, our system may be a superior, cost-effective
method that includes a minimal circuit, without requir-
ing additional, substantial equipment. The system is also
flexible, allowing the use of either HD or DUF, as neces-
sary, without eliminating the primary function of blood
concentration.
This system also provides fewer dilutional effects,
because the DUF experiment resulted in a lower Hct
level than that obtained via ALTHD, at all settings. In
general, ECC blood dilution may occur as the result of a
large priming volume; therefore, vascular permeability
can be easily enhanced.10,11 The risk of hypovolemia,
caused by fluid leakage, may be increased by further
blood dilution.12,13 Thus, the potential of avoiding dilu-
tional effects during actual surgeries is beneficial.
However, additional experimental work is required to
confirm these in vitro observations.
Considering the proposed clinical applicability of our
circuit, we believe that the proposed DUF/HD circuit-
switching technique has value as it allows clinicians to be
prepared for any eventuality; it can be available at all
times due to the use of a common hemoconcentration
circuit. In most circumstances, DUF is presumed to be
more efficient than HD at removing inflammatory
cytokines.14,15 Nakayama et al.16 demonstrated that DUF
is also efficient for the management of fluid and electro-
lyte balance during cardiopulmonary bypass, although it
is inferior to HD in these aspects. These considerations
suggest that DUF is normally an appropriate and suffi-
cient method.17
However, there are cases where the rapid correction of
potassium levels is necessary and where the availability
122 Perfusion 29(2)
Figure 4.  Efficiency curves comparing dilutional ultrafiltration (DUF) and hemodialysis (HD). The DUF data plotted as triangles and
HD data plotted as squares.
of this cost-effective circuit would benefit patients, such
as in cases where the potassium level increase is the result
of the rewarming of a hypothermia victim.18,19 Similarly,
rapid correction may be required in patients where a
high potassium level is intentionally maintained, such as
in repeat surgeries involving coronary artery bypass
grafts and in internal mammary artery grafts.20 Rapid
corrections are required in these instances because the
development of hyperkalemia, in these situations, is
more likely to trigger fatal arrhythmias;21 therefore, these
procedures should have the ability to rapidly adjust
potassium levels. Our circuit has value in terms of being
able to rapidly switch to ALTHD, as needed. In addition,
this alternative circuit also has value because of the
absence of a dilutional effect. This simple, low-cost con-
cept is suggested for applications in which the patients
have vascular hyperpermeability, such as dialysis, and in
pediatric patients.22-25 Finally, this novel circuit can be
used to develop an “ECC plan” through enhancement of
the rapid correction of electrolyte imbalances.
Conclusion
An ALTHD method, using a hemoconcentrator, was
successfully used to correct blood electrolytes in an in
vitro ECC loop. The efficiency was similar to that
observed with traditional DUF, but the simplicity of the
proposed circuit has the potential to increase the rapid-
ity of correcting electrolyte levels. Therefore, we believe
that this simple, minimal cost method of hemoconcen-
tration circuit switching has the potential to be rou-
tinely employed for rapidly correcting electrolyte
balances.
Declaration of conflicting interest
The author declares that there is no conflict of interest.
Funding
This research received no specific grant from any fund-
ing agency in the public, commercial or not-for-profit
sectors.
Tagaya et al. 123
References
  1. Roe BB, Hutchinson JC, Fishman NH, Ullyot DJ, Smith
DL. Myocardial protection with cold, ischemic, potas-
sium-induced cardioplegia. J Thorac Cardiovasc Surg
1977; 73: 366–374.
  2. Kyösola K, Chambers D, Canković-Darracott S,
Braimbridge MV. St. Thomas’ Hospital cardioplegia for
myocardial preservation during prolonged aortic cross-
clamping. Ann Chir Gynaecol 1985; 74: 111–117.
 3. Buckberg GD, Beyersdorf F, Allen BS, Robertson JM.
Integrated myocardial management: background and ini-
tial application. J Card Surg 1995; 10: 68–89.
  4. Bruemmer-Smith S, Avidan MS, Harris B, et al. Glucose,
insulin and potassium for heart protection during cardiac
surgery. Br J Anaesth 2002; 88: 489–495.
  5. Bethune DW, McKay R. Paradoxical changes in serum-
potassium during cardiopulmonary bypass in association
with non-cardioselective beta blockade. Lancet 1978; 2:
380.
  6. Kopman EA, Ramirez-Inawat RC. Persistent electrome-
chanical cardiac arrest following administration of cardi-
oplegic and glucose-insulin-potassium solutions. Anesth
Analg 1980; 59: 69–71.
  7. Khoo MS, Braden GL, Deaton D, et al. Outcome and com-
plications of intraoperative hemodialysis during cardio-
pulmonary bypass with potassium-rich cardioplegia. Am
J Kidney Dis 2003; 41: 1247–1256.
 8. Miyahara K, Maeda M, Sakurai H, Nakayama M,
Murayama H, Hasegawa H. Cardiovascular surgery in
patients on chronic dialysis: effect of intraoperative hemo-
dialysis. Interact Cardiovasc Thorac Surg 2004; 3: 148–152.
  9. Fukumoto A, Yamagishi M and Doi K, et al. Hemodia
filtration during cardiac surgery in patients on chronic
hemodialysis. J Card Surg 2006; 21: 553–558.
10. Kirklin JK. Prospects for understanding and eliminating
the deleterious effects of cardiopulmonary bypass. Ann
Thorac Surg 1991; 51: 529–531.
11. Verrier ED, Morgan EN. Endothelial response to car-
diopulmonary bypass surgery. Ann Thorac Surg 1998; 66:
S17–19; discussion S25–28. Review.
12. Toraman F, Evrenkaya S, Yuce M, et al. Highly posi-
tive intraoperative fluid balance during cardiac surgery
is associated with adverse outcome. Perfusion 2004; 19:
85–91.
13. Ohqvist G, Settergren G, Bergstrom K, Lundberg S.
Plasma colloid osmotic pressure during open-heart sur-
gery using non-colloid or colloid priming solution in the
extracorporeal circuit. Scand J Thorac Cardiovasc Surg
1981; 15: 251–255.
14. De Vriese AS, Colardyn FA, Philippé JJ, Vanholder RC, De
Sutter JH, Lameire NH. Cytokine removal during contin-
uous hemofiltration in septic patients. J Am Soc Nephrol
1999; 10: 846–853.
15. Bouman CS, Oudemans-van Straaten HM, Schultz MJ,
Vroom MB. Hemofiltration in sepsis and systemic inflam-
matory response syndrome: the role of dosing and timing.
J Crit Care 2007; 22: 1–12.
16. Nakayama Y, Sakata R, Ura M, Miyamoto T-A. Coronary
artery bypass grafting in dialysis patients. Ann Thorac
Surg 1999; 68: 1257–1261
17. Ko W, Kreiger KH, Isom OW. Cardiopulmonary bypass
procedures in dialysis patients. Ann Thorac Surg 1993; 55:
677–684.
18. Lim M, Linton RA, Band DM. Rise in plasma potassium
during rewarming in open-heart surgery. Lancet 1983;
291: 241–242.
19. Ivens D, Camu F. Sudden hyperkalemia during cardio-
pulmonary bypass with hypothermic cardiac arrest in an
infant. J Cardiothorac Vasc Anesth 1996; 10: 258–260.
20. Ramanathan T, Shekar P, Fitzgerald D, et al. Systemic
hyperkalemia for cardiac arrest on CPB with or without
cross-clamping. J Card Surg 2008; 23: 611–613.
21. El-Sherif N, Turitto G. Electrolyte disorders and
arrhythmogenesis. Cardiol J 2011; 18: 233–245.
22. Crosbie WA, Snowden S, Parsons V. Changes in lung capil-
lary permeability in renal failure. Br Med J 1972; 4: 388–390.
23. Ilson BE, Bland PS, Jorkasky DK, et al. Intraoperative
versus routine hemodialysis in end-stage renal disease
patients undergoing open-heart surgery. Nephron 1992;
61: 170–175.
24. Seghaye MC, Grabitz RG, Duchateau J, et al. Inflammatory
reaction and capillary leak syndrome related to cardiopul-
monary bypass in neonates undergoing cardiac opera-
tions. J Thorac Cardiovasc Surg 1996; 112: 687–697
25. Fukumura F, Kado H, Imoto Y, et al. Usefulness of low-
priming-volume cardiopulmonary bypass circuits and
dilutional ultrafiltration in neonatal open-heart surgery.
J Artif Organs 2004; 7: 9–12.