Vous êtes sur la page 1sur 8

Journal of Clinical Apheresis 28:310 (2013)


Therapeutic Plasma Exchange: A Technical and

Operational Review
Andre A. Kaplan1,2,3*
John Dempsey Hospital
University of Connecticut Health Center
UConn Dialysis Center

Therapeutic plasma exchange (TPE) is an extracorporeal blood purification technique designed for the removal
of large molecular weight substances. Examples of these substances include pathogenic autoantibodies, immune
complexes, cryoglobulins, myeloma light chains, endotoxin and cholesterol containing lipoproteins. The basic
premise of the treatment is that removal of these substances will allow for the reversal of the pathologic proc-
esses related to their presence. This review will cover the techniques for performing TPE, the kinetics of the re-
moval of large molecules from the plasma and the benefits and risks of the different types of replacement fluids.
J. Clin. Apheresis 28:310, 2013. V C 2013 Wiley Periodicals, Inc.

Key words: therapeutic plasma exchange; plasmapheresis; apheresis

INTRODUCTION AND RATIONALE The removal of pathogenic autoantibodies offers

an example. If one considers that the natural half
Therapeutic plasma exchange (TPE) is an extracorpor- life of IgG is approximately 21 days [4] and assum-
eal blood purification technique designed for the removal ing that an immunosuppressive agent could immedi-
of high-molecular-weight substances. Examples of these ately halt production (unlikely), the serum levels
substances include pathogenic autoantibodies, immune would still be 50% of the initial values for at least
complexes, cryoglobulins, myeloma light chains, endo- 21 days after initiating therapy. Such a delay might
toxin, and cholesterol containing lipoproteins. The basic be unacceptable in the presence of a very aggressive
premise of the treatment is that the removal of these sub- autoantibody such as that involved with Goodpas-
stances will allow for the reversal of the pathologic proc- tures syndrome.
esses related to their presence. Other potential benefits
for TPE include an unloading of the reticuloendothelial GENERAL GUIDELINES FOR PRESCRIBING TPE
system [1], stimulation of lymphocyte clones to enhance
cytotoxic therapy [2], and the possibility of reinfusing To prescribe TPE in a rationale manner, the amount of
large volumes of plasma without the risk of intravascular plasma to be exchanged must be determined in relation to
volume overload [3]. the patients estimated plasma volume (EPV). A simple
For TPE to be a rational choice as a blood purifica- means of estimating, the EPV can be calculated from the
tion technique, at least one of the following conditions patients weight and hematocrit using the formula [5]:
should be met:
EPV 50:0653wt kg312Hct (1)
1. The substance to be removed is sufficiently large
(>15,000 Da) so as to make other, less expensive
purification techniques unacceptably inefficient (i.e., This formula is one of many that have been pro-
hemofiltration or high flux dialysis). posed [69] and it is offered in this volume because it
2. The substance to be removed has a comparatively directly relates the plasma volume estimate to the
prolonged half-life, so that extracorporeal removal
provides a therapeutically useful period of dimin- *Correspondence to: Andre Kaplan. E-mail: kaplan@nso.uchc.edu
ished serum concentration.
Received 26 December 2012; Accepted 15 January 2013
3. The substance to be removed is acutely toxic and/or
Published online 19 February 2013 in Wiley Online Library
resistant to conventional therapy, so that the rapidity (wileyonlinelibrary.com).
of extracorporeal removal is clinically indicated. DOI: 10.1002/jca.21257
C 2013 Wiley Periodicals, Inc.
4 Kaplan

patients hematocrit, it allows for easy bedside calcula- In general, high-molecular-weight substances (immu-
tion and, in clinical use, it has provided a reliable pre- noglobulins, cholesterol-containing lipoproteins, and
diction of overall treatment results [9,10]. cyroglobulins) are only slowly equilibrated between
their extravascular and intravascular distribution. Thus,
removal during a single treatment is essentially limited
to which is in the intravascular compartment and the
amount of plasma to be exchanged to provide a given
reduction in pretreatment levels can be determined by
application of first-order kinetics using the formula:
X1 5X0e 2Ve =EPV (2)
where X1 equals the final plasma concentration, Xoe
equals the initial concentration, Ve equals the volume
exchanged, and EPV equals the patients EPV. The
relationship is shown in Figure 1. For example, if the
volume exchanged (Ve) is equal to the patients EPV,
Fig. 1. Relation of volume exchanged, EPV, and percent reduction pretreatment values will be lowered by 63%. If the
in initial concentration for high-molecular-weight substances plasma exchanged is equal to 1.4 times the EPV, the pre-
removed during TPE. For example, if the volume exchanged (Ve) is
equal to the patients EPV, then Ve/EPV will be equal to one and
treatment levels will be lowered by 75%. As shown in
pretreatment values will be lowered by 63%. If the plasma Figure 1, increasingly voluminous exchanges during a
exchanged is equal to 1.4 times the EPV, the pretreatment levels single treatment yield a progressively smaller reduction
will be lowered by 75%. As shown in the figure, increasingly volu- in pretreatment levels. Given that overly voluminous
minous exchanges during a single treatment yield a progressively exchanges will inefficiently increase the duration and
smaller reduction in pretreatment levels. For most indications, each
treatment should provide an exchange volume equaling 11.4 times costs of the treatment (i.e., replacement fluid and nursing
the EPV. Reproduced with permission from Kaplan AA, A Practical time), for most indications each treatment should provide
Guide to Therapeutic Plasma Exchange, copyright Andre Kaplan. an exchange volume equaling 11.4 times the EPV.

TABLE I. Distribution and Metabolism of Plasma Proteinsa

Concentration Percent Fractional turnover Half life

Protein (mg/mL) M.W. 3 103 Da intravascular rate (% day) (days)
Normal Physiology
IgG (except IgG3 subclass) 12 150 45 7 22
IgG3 0.7 150 64 17 7
IgMa 0.9 950 78 19 5
IgA 2.5 160 42 25 6
IgD 0.02 175 75 37 2.8
IgE 0.0001 190 45 94 2.5
Albumin 45 66 44 11 17
C3 1.4 240 67 41 2
C4 0.5 200 66 43 2
Fibrinogen 34 340 81 24 4.2
Factor VIII 0.1 100340 71 150 0.6
Antithrombin III 0.2 5658 45 55 2.4
Lipoprotein cholesterol 1.52.0 1,300 >90 35
Pathologic conditions
Macroglobulinemia, IgM 50130 950 89 25* 5.9
Bence-Jones protein 410 3 10225 <50 # #
Endotoxin 325 3 1027 1002,400* >50 @ @
Immune complexes * >300* >50 @ @
TNF 35 3 1027 50 (Trimer) <50 620 Min
Values listed are averaged from those reported in the literature. Removal of a substance during a single TPE treatment will be limited to
which is intravascular. Substances with substantial extravascular distribution will require several consecutive TPE treatments to decrease
total body burden. Those substances with short half lives (high turnover rate) will have a rapid return to pre TPE levels unless production
rates can be slowed by concomitant therapy.
#Highly dependent on degree of renal function, half life greatly increased with renal failure.
@Half life will be variable and dependent on the clearing capabilities of the reticulo-endothelial system.
*Highly variable or poorly defined.
Reproduced from: Kaplan AA: A Practical Guide to Therapeutic Plasma Exchange. Blackwell Science, Malden, MA, 1999 copyright Andre Kaplan.

Journal of Clinical Apheresis DOI 10.1002/jca

Therapeutic Plasma Exchange 5

quantitative means, such as with a specific autoanti-

body, then the treatment schedule should be designed
to achieve a substantial lowering of that substance
using the kinetic considerations outlined above and tak-
ing into account the substances rate of increase in the
postpheresis period. If treatments are performed with-
out identification of the offending agent, then the phy-
sician is dependent on empiric treatment schedules
derived from the literature.


The basic tenets of immunoglobulin kinetics can be
gleaned from the results of experiments in which isoto-
pically labeled immunoglobulins have been infused
into humans [4]. These experiments have demonstrated
Fig. 2. Progressive decline in IgG levels after three consecutive
that immunoglobulins have relatively long half lives,
TPE treatments equaling one plasma volume each. Intertreatment
increases between treatments represent a combination of extravascu- approaching 21 days for IgG and 5 days for IgM,
lar to intravascular re-equilibration and a variable amount of new immunoglobulins have a substantial extravascular dis-
IgG synthesis. Reproduced from Kaplan AA: A Practical Guide to tribution, approximately 60% for IgG and 20% for
Therapeutic Plasma Exchange, copyright Andre Kaplan.
IgM, and immunoglobulins exhibit an intravascular to
extravascular equilibration that is relatively slow,
It must be noted that the observed decline in serum approximating 13% per hour [4,17,24].
concentrations of a given substance after a single TPE Considering the relatively long half lives of immuno-
treatment bears little relationship to the absolute globulins, the use of immunosuppressive agents which
decrease in total body load. Soon after the initial act only by depressing antibody synthesis cannot be
reduction in serum levels, there will be a partial expected to substantially lower the levels of a patho-
rebound. One component of this rebound represents a genic autoantibody for at least several weeks, even if
renewed synthesis of the substance, whereas another production is completely blocked. Hence, the basic ra-
component is owing to the extravascular to intravascu- tionale for their removal by extracorporeal means. As
lar redistribution of the substance. Many high-molecu- detailed in the previous section, the relatively slow re-
lar-weight substances, such as the immunoglobulins, equilibration between the extravascular and the intravas-
have a substantial extravascular distribution Table I cular compartment allows the use of first-order kinetics
[1122]. After the rapid reduction in intravascular con- governing removal rates from a single compartment
centrations provided by the TPE treatment, the extrava- (i.e., the intravascular space). The calculations require
scular distribution of the substance will re-equilibrate that the volume exchanged (Ve) to be related to the vol-
with the intravascular space. Considering that the ume of distribution of the substance from which it can
extravascular distribution of a given substance cannot be be removed during the procedure, a volume which is
removed until it has entered the intravascular space, and limited to the intravascular space and, for practical pur-
considering that the extravascular to intravascular reequi- poses, can be considered to be equal to the patients
libration of a high-molecular-weight substance will be EPV. An example of this exponential decline is shown
relatively slow (approximately, 13% per hour), several in Figure 1 from the previous section. If the fraction Ve/
consecutive treatments, separated by 2448 h each, will EPV equals 0.7, the expected decrease will be 50%, if
have to be performed to remove a substantial percentage Ve equals EPV the decrease should be 63% and when
of the total body burden. An example of the progressive Ve/EPV equals 1.4, the decrease should be 75%. Clinical
reduction in serum levels of an immunoglobulin is shown validation of these predictions has been demonstrated
in Figure 2 [23]. In general, if production rates (resynthe- for each of the major immunoglobulin classes [5,10].
sis) are modest (i.e., slowly forming antibody), at least It must be noted that the observed decline in immu-
five separate treatments over a 7- to 10-day period will noglobulin levels after a single plasma exchange bears
be required to remove 90% of the patients initial total little relationship to the absolute decrease in total body
body burden [2327]. If production rates are high (i.e., load. In effect, after a given TPE treatment, the
rapidly forming antibody, complement components), extravascular distribution of an immunoglobulin will
additional treatments may be required [28,29]. begin to enter the vascular space yielding a post-treat-
In conclusion, a rational approach to prescribing ment increase which will begin to level after 2448 h.
plasma exchange can be considered as follows: If the After this re-equilibration, there will be a further op-
substance to be removed is measurable by reliable portunity for substantial removal of the
Journal of Clinical Apheresis DOI 10.1002/jca
6 Kaplan

immunoglobulin by a subsequent TPE treatment. Using TECHNIQUE

a mathematical model, Roberts et al. [24] have studied
Traditionally, plasma exchange was performed with
the kinetics of multiple plasma exchanges in rabbits.
centrifugation devices used in blood blanking proce-
Their model predicts a 74% decline in total body
dures. These devices offer the advantage of allowing
load after three daily exchanges equaling one plasma
for selective cell removal (cytapheresis) [30]. Plasma
volume each, the observed level was 69%. In close
exchange can also be performed with a highly permea-
agreement with this animal data are our own obser-
ble filter and standard dialysis equipment, a technique
vations obtained from nine series of treatments in
which is often referred to as membrane plasma
seven patients. As shown in Figures 2 and 3, the
separation (MPS) [31]. A detailed review of the
expected re-equilibration after three daily exchanges
available removal systems has been provided by
yields an approximate 70% decline in total body
Sowada et al. [32].
load for IgG and an approximate 80% decline for
IgM [23]. It is worth noting that similar predictions
can be made for other high-molecular-weight sub-
stances; however, extravascular to intravascular re- Centrifugal systems for plasma exchange utilize G
equilibration constants and overall serum half lives forces to separate the plasma into its different compo-
will have to be known to perform the calculations nents. In increasing order of density (specific gravity,
(Table 2 of Ref. 26]) [26]. SG), whole-blood constituents are layered into plasma
It should be kept in mind that the above-referenced (SG, 1.0251.109), platelets (1.040), lymph (1.070),
calculations and the observed results describe a best- granulocytes (1.0871.092), and red cells (1.093
case scenario concerning immunoglobulin removal. In 1.096). Separation of the plasma can be either intermit-
essence, these data are for an entire immunoglobulin tent or continuous. Newer devices utilize a continuous
class and not for any specific antibody. Although cor- flow system in which the whole blood is processed in
relation between total IgG levels and pathogenic anti- an ongoing, online manner allowing the extracorporeal
bodies may be tightly correlated, such as with the volume to be limited to that which is necessary for the
relatively slowly produced antibody in myasthenia blood tubing and the centrifugal receptacle. Most of
gravis [10], in other autoimmune diseases, the rate of the newer devices also offer several user-friendly mod-
autoantibody production may greatly exceed that of the ifications, including means for adjusting the amount of
total immunoglobulin class. Such phenomenon has citrate anticoagulation, automated control of blood
been documented for certain cases of Goodpastures flow and plasma separation and volume regulation of
syndrome where anti-glomerular basement membrane the replacement fluid.
activity will be predictably lowered by a given plasma
exchange treatment but for which the intertreatment FILTRATION (MPS)
increases in serum levels are too rapid to be compatible
Separation of plasma from the bloods cellular com-
with a simple re-equilibration of extravascular stores
ponents can also be accomplished by filtration though
[28]. The same discordance has been documented in
a highly permeable membrane. This methodology sepa-
systemic lupus erythematosus, where, after extracorpor-
rates the blood into its cellular and noncellular compo-
eal removal by plasma exchange, anti-DNA-binding ac-
nents by subjecting it to sieving through a membrane
tivity may be produced at a far grater rate than that of
whose pores allow the plasma proteins to pass but
the entire IgG class [29]. Thus, a 7080% absolute
which retain the larger cellular elements within the
decrease in a pathogenic autoantibody requires at least
blood path. Pore sizes of these membranes are usually
three plasma exchange treatments, and may require a
0.6 m or less, and thus easily rejecting the smallest cel-
far more intensive treatment schedule if production
lular component, the platelets (3 m).
rates cannot be adequately controlled by the concomi-
Anticoagulation: Most TPE treatments performed
tant immunosuppressive medications.
with centrifugal devices use citrate anticoagulation.
Another potentially important factor which may ne-
Although this technique can be modified for use with
cessitate a more extensive prescription of plasma
an MPS system [32,33], most MPS treatments are per-
exchange is the possibility of a post-treatment stimula-
formed with heparin.
tion of pathogenic clones. Schroeder et al. [2] have
reviewed the animal data and anecdotal human reports
which suggest a stimulation of pathogenic antibody
production after extracorporeal removal. These data Regardless of the technique employed, TPE will
underscore the importance of adding concomitant normally require some form of anticoagulation to avoid
immunosuppressive therapy to any plasma exchange clotting within the extracorporeal circuit. For centrifu-
schedule designed to control an aggressive autoimmune gal techniques, this is often provided by citrate infu-
disease. sions which bind ionized calcium in the extracorporeal
Journal of Clinical Apheresis DOI 10.1002/jca
Therapeutic Plasma Exchange 7

circuit such that the coagulation cascade is impeded. patients plasma (18 vs. 22 torr) but at least one of the
The ionized calcium level returns toward its original patients studied had a condition in which oncotic
level as the blood is returned to the intravascular com- pressure would be abnormally high (Waldenstroms
partment where there are substantial stores of ionized macroglobulinemia) [37]. In contrast, Chopek and
calcium and where the citrate will be metabolized. McCullough [11] have demonstrated a slight hemodilu-
Nonetheless, signs of hypocalcemia ranging from cir- tion effect occurring after replacement with 5% albu-
cumoral paresthesias to QT prolongation of the electro- min, suggesting that, in clinical use, this solution is
cardiogram represent the most commonly reported slightly hyperoncotic and may result in mild intravascu-
secondary effects associated with TPE procedures lar expansion. McLeod et al. [38] have studied the use
[34,35]. Thus, the goal of citrate anticoagulation is to of partial albumin replacement, using a replacement
provide enough to prevent clotting within the extracor- schedule involving 750 mL of saline followed by 1,250
poreal circuit while limiting the degree of systemic mL of 5% albumin. They found this combination to be
hypocalcemia resulting from the intravascular infusion well tolerated and no patient developed hypotension or
of the citrate. In an indepth review of the subject, Hes- peripheral edema, but the exchanges were limited to 2
ter et al. [36] recommend that citrate infusions be lim- L, a rather modest volume for most TPE prescriptions.
ited to between 1.0 and 1.8 mg/kg/min. Maintaining
levels at, or below, 1.8 mg/kg/min can be roughly esti-
mated by choosing a given ratio of citrate solution to Electrolyte composition
whole blood and limiting the blood flows to a multiple of Five percent human serum albumin is isosmotic to
the patients body weight. Clearly, however, these recom- plasma, contains no preservatives, and is characterized
mendations must be considered as general guidelines by a sodium level of approximately 145 1 15 mmol/L
because of the extreme variability of a given patient to and a potassium level lower than 2 mmol/L [39]. The
metabolize citrate, the possibility of pretreatment abnor- relative lack of potassium can result in a 25% reduc-
malities in ionized calcium and the variable content of ci- tion of serum potassium levels in the immediate post-
trate in the replacement solutions (fresh frozen plasma, pheresis period, leading to a risk of hypokalemic
FFP, can contain up to 14% citrate by volume). If symp- arrhythmia (a situation most likely to occur in patients
toms do occur, the same authors suggest calcium replace- taking digitalis preparations). This type of postpheresis
ment with 10 mL of 10% calcium gluconate infused over hypokalemia can be avoided by adding 4 mmol of po-
15 min, approximately half way through the procedure. tassium to each liter of 5% albumin.
Others believe that citrate toxicity can be reasonably well Albumin solutions are contaminated with between 4
controlled with the oral administration of calcium tablets and 24 mmol/L of aluminum [40,41] and massive replace-
during the procedure, reserving intravenous calcium ment with albumin may result in aluminum toxicity.
replacement only for those who develop symptoms.

REPLACEMENT FLUIDS: Anaphylactic reactions

ALBUMIN Human serum albumin consists of 96% albumin and
trace amounts of a- and b-globulins and, as opposed to
General comments
FFP, anaphylactoid reactions are rare and may be asso-
Albumin is the most commonly used replacement ciated with the formation of antibodies to polymerized
fluid in the United States and, when compared to FFP, albumin created by heat treatment or stabilization with
has the advantage of lacking viral transmission and sodium caprylate [39,42,43]. Recent reports have sug-
possessing a decreased risk of anaphylactoid reactions. gested that patients taking angiotensin converting
Disadvantages include a post-treatment coagulopathy enzyme inhibitors may also be prone to increased risk
related to the removal of clotting factors and a net loss of atypical or hypotensive reactions to albumin
of immunoglobulins. A 5% concentration of albumin [44,45] (see sections on Complications).
will provide a reasonable replacement of the oncotic
pressure removed with the patients plasma (see
below). Some centers prefer to dilute the albumin to Depletion coagulopathy
approximately 3.5%, a solution which is hypooncotic When albumin is used as the sole replacement fluid,
to the plasma which is being removed and may render there is a depletion of all coagulation factors. After a
the patient more prone to hypotension. single plasma exchange, prothrombin time increases
approximately 30% and partial thromboplastin time
doubles [46]. Although there is variability among
Colloid oncotic pressure patients, especially when clotting factor production
In a study that measured colloid oncotic pressure, 5% may be compromised (liver failure) partial thrombo-
albumin was found to be slightly hypooncotic to the plastin time and thrombin time return toward the
Journal of Clinical Apheresis DOI 10.1002/jca
8 Kaplan

normal range in approximately 4 h, whereas prothrom- 21% [34] (Table 6 of Ref. 34]). Symptomatology may
bin time normalizes in 24 h [47]. When multiple treat- vary and include fever, rigors, urticaria, wheezing, and
ments are performed over a short period (i.e., three or hypotension [53,54]. These reactions are among the
more treatments per week), the depletion in clotting most serious encountered during TPE procedures and
factors is more pronounced and may require several the uncommon reports of TPE-related deaths and are
days for spontaneous recovery [34,47]. Under these most often associated with FFP replacement [5558].
conditions, the risks of hemorrhage can be minimized As these reactions may involve the formation of kinins,
with a partial replacement of FFP given toward the end the use of ACE inhibitors should be avoided in patients
of the procedure. undergoing TPE.
Because of the relatively high incidence of these
reactions, patients undergoing massive replacement
Immunoglobulin depletion with FFP (for TTP or HUS) are commonly pretreated
When albumin is used as the replacement fluid, re- with 50 mg of diphenhydramine. In those patients who
moval of immunoglobulins and complement may pre- have already demonstrated a sensitivity to FFP, we rec-
dispose patients to high rates of infection. One plasma ommend oral dosing of 50 mg of prednisone, given 13,
volume exchange will result in a 60% reduction in se- 7, and 1 h before the treatment, combined with 50 mg
rum immunoglobulin levels and a net 20% reduction in of diphenhydramine and 25 mg of ephedrine, given 1 h
total body immunoglobulin stores [48]. Multiple treat- before the treatment [42]. In the event of a severe life-
ments over short periods, especially when associated threatening reaction (laryngeal edema, etc.), 0.30.5
with immunosuppressive agents, will yield a substantial mL of epinephrine (solution, 1:1,000) should be avail-
decrease in immunoglobulin levels that may persist for able for subcutaneous administration.
several weeks [27,48]. If serious infection occurs soon
after a series of TPE treatments, a one-time infusion of
immunoglobulin (IVIG) at 100400 mg/kg will recon- Citrate toxicity
stitute normal immunoglobulin levels. FFP contains approximately 14% citrate by volume
and large infusions of FFP may lead to symptoms of
hypocalcemia and metabolic alkalosis. Symptoms of
Risk of viral transmission
hypocalcemia can be avoided with the prophylactic
Albumin preparations are treated with heat and are replacement of calcium. Metabolic alkalosis, most
considered to be devoid of transmissible virus [39]. common in patients with severe renal failure, may
The same claims had been made for intravenous immu- require concomitent hemodialysis [34,59]. In this
noglobulins [50], but an outbreak of hepatitis C from regard, those patients requiring TPE with FFP replace-
contaminated IVIG has been documented [51] and new ment who are also undergoing hemodialysis (TTP/
methodologies for avoiding viral transmission from HUS, Goodpastures syndrome post-biopsy, etc.)
IVIG preparations have been initiated [52]. should be hemodialyzed after the TPE treatment to
facilitate correction of the citrate-induced alkalemia.
General comments Risk of viral transmission
FFP contains all the noncellular components of nor- Risk of viral transmission during plasma exchange
mal blood and does not lead to postpheresis coagulop- is directly related to replacement with FFP. The current
athy nor immunoglobulin depletion. FFP is also incidences of transfusion-acquired viral infections have
considered essential for the treatment of thrombotic declined substantially from the early 1980s and are
thrombocytopenic purpura (TTP) as TPE for this indica- currently estimated as: 12 per 1,000,000 units for HIV
tion may be most useful as a means of providing a miss- and hepatitis C and 1 per 200,000500,000 units for
ing serum factor [3]. Disadvantages include Hepatitis B [6062]. It should be noted that during a
anaphylactoid reactions (most often mild, but can be life single plasma volume exchange with FFP (approxi-
threatening), citrate toxicity and a small, but persistent mately, 3 L), 1015 U, obtained from an equal number
risk of viral transmission. Because of these potential of donors, are used.
problems, FFP should be avoided except for the treat-
ment of TTP/HUS or when hemorrhagic risks are great.
Product shortages and rising costs may necessitate
Anaphylactoid reactions the use of nonprotein-containing solutions as replace-
Anaphylactoid reactions to FFP are common and ment fluids for TPE. Recently, Owen and Brecher [63
have been reported to occur with an incidence of up to 65] have investigated the use of hydroxyethyl starch
Journal of Clinical Apheresis DOI 10.1002/jca
Therapeutic Plasma Exchange 9

(Hespan) as a partial or full replacement for albumin. 11. Chopek M, McCullough J. Protein and biochemical changes
This starch-based colloid solution is biochemically sim- during plasma exchange. In: Ulmas J. Berkman E, editors. Ther-
apeutic Hemapheresis: A Technical Workshop. Washington,
ilar to glycogen rendering it unlikely to engender DC: American Association of Blood Banks, 1980. pp 1352.
immune reactions. In one study, seven patients 12. Morgenthaler JJ, Nydegger UE. Synthesis, distribution and ca-
received 1,000 mL of 6% hetastarch as part of their tabolism of human plasma proteins in plasma exchange. Int J
replacement fluid during 33 procedures, whereas a fur- Artif Organs 1984;7:2734.
ther 42 patients received 1,000 mL of 3% hetastarch 13. Thornton CA, Griggs RC. Plasma exchange and intravenous im-
munoglobulin treatment of neuromuscular disease. Ann Neurol
during 289 procedures [64]. In those receiving 3% 1994;35:260268.
hetastarch, the blood pressure and pulse remained sta- 14. Bowman BH. Hepatic Plasma Proteins: Mechanisms of Function
ble for 97.3% of the procedures (280 treatments). Two and Regulation. San Diego: Academic Press, Inc., Harcourt
patients receiving the 6% solution and one in the 3% Brace Jovanovich, 1993.
group complained of severe transient back and head 15. Herbert LA. The clearance of immune complexes from the cir-
culation of man and other primates. Am J Kidney Dis
pain during hetastarch infusion. Total protein decreased 1991;27:352361.
in both groups but only one patient reported slight pe- 16. Barth WF, Wochner D, Waldmann TA, Fahey JL. Metabolism of
ripheral edema after two procedures, suggesting a mod- human gamma macroglobulins. J Clin Invest 1964;43:10361048.
est decline in serum oncotic pressure. The authors 17. Russell, JA, Fitzharris, BM, Corringham, R, Darcy DA, Powles
concluded that 3% hetastarch was a safe and cost- RL. Plasma exchange v peritoneal dialysis for removing Bence
Jones protein. Br Med J 1978;2:1397.
effective partial replacement for albumin during TPE. 18. Fisher Jr, CJ, Agosti JM, Opal SM, Lowry SF, Balk RA, Sadoff
Kinetic modeling of the wash-out of starch used as a JC, Abraham E, Schein RM, Benjamin E. Treatment of septic
replacement fluid demonstrated that relatively little re- shock with the tumor necrosis factor receptor: Fc fusion protein.
sidual starch remains compared to the total amount N Engl J Med 1996;334:16971702.
infused [65]. As an example, if six successive 4-L 19. May ME, Mintz PD, Gray LS. Multicompartment analysis of
the effects of plasmapheresis: application to lipid kinetics in
exchanges were performed using only starch as the humans. Am J Clin Pathol 1989;91:688694.
replacement solution, only 4.1 L of starch (4 L 3 6 20. Carter PW, Cohen HJ, Crawford J. Hyperviscosity syndrome in
exchanges 3 17% 5 4.1 L) would be expected to association with kappa light chain myeloma. Am J Med
remain in the circulation despite a total infusion of 24 1989;86:591601.
L (4 L 3 6 exchanges 5 24 L) of starch. 21. Misiani R, Tiraboschi G, Mingardi G, Mecca G. Management of
myeloma kidney: an anti-light chain approach. Am J Kidney
Dis 1987;10:2833.
REFERENCES 22. Schetz M, Ferdinande P, Van den Berghe G, Verwaest C,
Lauwers P. Removal of pro inflammatory cytokines with renal
1. Lockwood CM, Worlledge S, Nicholas A, Cotton C, Peters DK.
replacement therapy: sense or non-sense? Intensive Care Med
Reversal of impaired splenic function in patients with nephritis
or vasculitis (or both) by plasma exchange. N Engl J Med
23. Kaplan AA. Towards a rational prescription of plasma
exchange: the kinetics of immunoglobulin removal. Semin Dial
2. Schroeder JO, Euler HH, Loffler H. Synchronization of plasma-
pheresis and pulse cyclophosphamide in severe systemic lupus
24. Roberts CG, Schindhelm K, Smeby LC, Farrell PC. Kinetic
erythematosus. Ann Intern Med 1987;107:344346.
analysis of plasma separation: use of an animal model. In:
3. Rock GA, Shumak KH, Buskard NA, Blanchette VS, Kelton Lysaght MJ, Gurland HJ, editors. Plasma Separation and Plasma
JG, Nair RC, Spasoff RA. Comparison of plasma exchange with Fractionation. Basel, Switzerland: Karger, 1983. pp 2538.
plasma infusion in the treatment of TTP. N Eng J Med
25. Charlton B, Schindhelm K, Farrell PC. Effect of extracorporeal
IgG removal on IgG kinetics. Trans Am Soc Artif Intern Organs
4. Cohen S, Freeman T. Metabolic heterogeneity of human gamma 1983;29:724729.
globulin. Biochem J 1960;76:475487. 26. Kellogg RM, Hester JP. Kinetics modeling of plasma exchange:
5. Kaplan AA. A simple and accurate method for prescribing intra and post-plasma exchange. J Clin Apheresis 1988;4:183
plasma exchange. Trans Am Soc Artif Intern Organs 187.
1990;36:M597M599. 27. Keller AJ, Urbaniak SJ. Intensive plasma exchange on the cell
6. Inkley SR, Brooks L, Krieger H. A study of methods for separator: effects on serum immunoglobulins and complement
the prediction of plasma volume. J Lab Clin Med 1955;45:841850. components. Br J Haematol 1978;38:531540.
7. Retzlaff JA, Newlon Tauxe W, Kiely JM, Stroebel CF. Erythro- 28. Swainson P, Robson JS, Urbaniak SJ, Keller AJ, Kay AB.
cyte volume, plasma volume and lean body mass in adult men Treatment of Goodpastures disease by plasma exchange and
and women. Blood 1969;33:649667. immunosuppression. Clin Exp Immunol 1978;32:233242.
8. Feldschuh J, Enson Y. Prediction of the normal blood volume: 29. Jones JV, Robinson MF, Parciany RK, Layfer LF, McLeod B.
relation of blood volume to body habitus. Circulation Therapeutic plasmapheresis in systemic lupus erythematosus:
1977;56:605612. effect on immune complexes and antibodies to DNA. Arthritis
9. Sprenger KBG, Huber K, Kratz, Henze E, Franz HE. Prediction Rheum 1981;24:11131120.
of patients plasma volume in plasma exchange therapy. In: 30. Gurland HJ, Lysaght MH, Samtleben W, Schmidt B. A compar-
Smeby LC, Jorstad S, Wideroe TE, editors. Immune and Meta- ison of centrifugal and membrane based apheresis formats. Int J
bolic Aspects of therapeutic Blood Purification Systems. Basel, Artif Organs 1984;7:3538.
Switzerland: Karger. 1986. pp 394402. 31. Gurland HJ, Lysaght MJ, Samtleben W, Schmidt B. Compara-
10. Kaplan AA, Halley SE. Plasma exchange with a rotating filter. tive evaluation of filters used in membrane plasmapheresis.
Kidney Int 1990;38:160166. Nephron 1984;36:173182.

Journal of Clinical Apheresis DOI 10.1002/jca

10 Kaplan
32. Sowada K, Malchesky PS, Nose Y. Available removal systems: 50. Haas A. Use of intravenous immunoglobulin in immunoregula-
state of the art. In: Nydegger UE, editor. Therapeutic Hemaphere- tory disorders. In: Stiehm ER, moderator, editor. Intravenous
sis in the 1990s. Current Studies in Hematology and Blood Trans- Immunoglobulins as Therapeutic Agents. Ann Intern Med
fusion, Vol. 57. Basel, Switzerland: Karger. 1990. pp 51113. 1987;107:367382.
33. Ilamathi E, Kirsch M, Moore B, Finger M. Citrate anticoagula- 51. Bjoro K, Froland SS, Yun Z, Samdal HH, Haaland T. Hepatitis
tion during plasmapheresis using standard hemodialysis equip- C infection in patients with primary hypogammaglobulinemia
ment. Semin Dial 1993;6:268. after treatment with contaminated immune globulin. N Engl J
34. Mokrzycki MH, Kaplan AA. Therapeutic plasma exchange: Med 1994;331:16071611.
complications and management. Am J Kidney Dis 1994;23:817 52. Schiff RI. Transmission of viral infections through intravenous
827. immune globulin (editorial). N Eng J Med 1994;331:1649
35. Olson PR, Cox C, McCullough J. Laboratory and clinical effects 1650.
of the infusion of ACD solution during platelet-pheresis. Vox
Sang 1977;33:7987. 53. Ring J, Messmer K. Incidence and severity of anaphylactoid
36. Hester JP, McCullough J, Mishler JM, Szymanski IO. Dosage reactions to colloid volume substitutes. Lancet 1977;1:466469.
regimens for citrate anticoagulants. J Clin Apheresis 54. Bambauer R, Jutzler GA, Albrecht D, Keller HE, Kohler M.
1983;1:149157. Indications of plasmapheresis and selection of different substitu-
37. Lasky LC, Finnerty EP, Genis L, Polesky HF. Protein and colloid tion solutions. Biomater Artif Cells Artif Organs 1989;17:927.
osmotic pressure changes with albumin and/or saline replacement 55. Huestis DW. Mortality in therapeutic haemapheresis (letter).
during plasma exchange. Transfusion 1984;24:256259. Lancet 1983;1:1043.
38. McLeod BC, Sassetti RJ, Stefoski D, Davis FA. Partial plasma 56. Aufeuvre JP, Morin-Hertel F, Cohen-Solal M, Lefloch A, Baudelot
protein replacement in therapeutic plasma exchange. J Clin J. Hazards of plasma exchange. In: Sieberth HG, editor. Plasma
Apheresis 1983;1:115118. Exchange. Stuttgart, Germany: FK Schattauer. 1980 pp 149157.
39. Finlayson JS. Albumin products. Semin Thromb Hemost 57. Aufeuvre JP, Mortin-Hertel F, Cohen-Solal M, Lefloch A, Baude-
1980;6:85120. lot J. Clinical tolerance and hazards of plasma exchanges: a study
40. Mousson C, Charhon SA, Ammar M, Accominotti M, Rifle G. of 6200 plasma exchanges in 1033 patients. In: Beyer JH, Bur-
Aluminum bone deposits in normal renal function patients after gerg H, Fuchs C, Nagel GA, editors. Plasmapheresis in Immunol-
long-term treatment by plasma exchange. Int J Artif Organs ogy and Oncology. Basel, Switzerland: Karger. 1982. pp 6577.
41. Milliner DS, Shinaberger JH, Shurman P, Coburn JW: Inadver- 58. Sutton DMC, Nair R, Rock G and the Canadian Apheresis
tent aluminum administration during plasma exchange due to Study Group. Complications of plasma exchange. Transfusion
aluminum contamination of albumin replacement solutions. N 1989;29:124127.
Engl J Med 1985;312:165167. 59. Pearl RG, Rosenthal MH. Metabolic alkalosis due to plasma-
42. Apter AJ, Kaplan AA. An approach to immunologic reactions pheresis. Am J Med 1985;79:391393.
with plasma exchange. J Allergy Clin Immunol 1992;90:119124. 60. Stramer SLGlynn SA, Kleinman SH, Strong DM, Caglioti S,
43. Stafford CT, Lobel SA, Fruge BC, Moffitt JE, Hoff RG, Fadel Wright DJ, Dodd RY, Busch MP; National Heart, Lung, and
HE. Anaphylaxis to human serum albumin. Ann Allergy Blood Institute Nucleic Acid Test Study Group. Detection of
1988;61:8588. HIV-1 and HCV infections among antibody-negative blood
44. Brecher ME, Owen Hg, Collins ML. Apheresis and ACE inhibi- donors by nucleic acidamplification testing. N Engl J Med
tors (letter). Transfusion 1993;33:963964. 2004;351:760768.
45. Owen HG, Brecher ME. Atypical reactions associated with use 61. Stramer SL. Current risks of transfusion-transmitted agents: a
of angiotensin-converting enzyme inhibitors and apheresis. review. Arch Pathol Lab Med 2007;31:702707.
Transfusion 1994;34:891894. 62. Dwyre DM, Fernando LP, Holland PV. Hepatitis B, hepatitis C
46. Kaplan AA, Halley SE. Plasma exchange with a rotating filter. and HIV transfusion-transmitted infections in the 21st century.
Kidney Int 1990;38:160166. Vox Sanguinis 2011;100:9298.
47. Chrinside A, Urbaniak SJ, Prowse CV, Keller AJ. Coagulation
abnormalities following intensive plasma exchange on the cell 63. Brecher ME, Owen HG, Bandarenko N. Alternatives to albumin:
separator. Br J Haematol 1981;48:627634. starch replacement for plasma exchange. J Clin Apheresis
48. Kaplan AA. Towards a rational prescription of plasma 1997;12:146153.
exchange: the kinetics of immunoglobulin removal (editorial). 64. Owen HG, Brecher ME. Partial colloid starch replacement for
Semin Dial 1992;5:227229. therapeutic plasma exchange. J Clin Apheresis 1997;12:8792.
49. Sultan Y, Bussel A, Maisonneuve P, Sitty X, Gajdos P. Poten- 65. Brecher ME, Owen HG. Washout kinetics of colloid starch as a
tial danger of thrombosis after plasma exchange in the treatment partial or full replacement for plasma exchange. J Clin Aphere-
of patients with immune disease. Transfusion 1979;19:588593. sis 1996;11:123126.

Journal of Clinical Apheresis DOI 10.1002/jca