Therapeutic

Fran#{231}oisMadore2,

Plasma
J. Michael

Exchange
and Hugh

in Renal
P. Brady SCIENTIFIC

Diseases1

Lazarus,

RATIONALE
some The potentially technique beneficial is used acmost exchange.

F. Madore, Department tal, Harvard

J.M. Lazarus, HR. Brady, Renal Division, of Medicine, Brigham & Womens HospiMedical School, Boston, MA

Table
tions

1 summarizes

ofplasma

frequently

to modulate

humoral

components

of the

H.P. Brady, Renal Section, Medical Service, BrocktonWest Roxbury Department of Veterans Affairs Medical Center, Harvard Center for the Study of Kidney Diseases, Boston, MA (J. Am. Soc. Nephrol. 1996; 7:367-386)

immune response and rapidly lower circulating titers of autoantibodies (e.g. , anti-glomerular basement membrane disease) or immune complexes (e.g. . cryoglobulinemia), while immunosuppressive therapy suppresses de novo antibody production ( 1 ). Plasma exchange has also been proposed as a useful adjunct
to chemotherapy for the removal of circulating immu-

ABSTRACT Plasma exchange has been used extensively for over 21/2 decades to treat a variety of renal diseases. In this article, the scientific rationale for therapeutic plasma exchange in primary and secondary glomerulonephritis, thrombotic thrombocytopenic purpura and hemolytic uremic syndrome, myeloma cast nephropathy, and allograft rejection are reviewed. The clinical studies that evaluate its efficacy are summarized, with special emphasis on the results of randomized controlled trials, when available. Consensus plasma exchange regimens are presented for diseases in which there is evidence to support its use. Key Words:
multiple Plasmapheresis, HUS/17P glomerulonephritis. vasculitis,

noglobulin and other

(Ig) or Ig components dysprotetnemlas for

(7).

in multiple In addition,

myeloma plasma

exchange

other thrombotic putative ulosclerosis)

is utilized than ig, such

as

thrombocytopenic toxic mediators (9). Infusion

the removal of components thrombotic factors (e.g. . in purpura) (8) or other (e.g. , focal segmental glomerof normal plasma may itself

have
normal evidence ponent anism

beneficial

effects,

independent

of removal

of ab-

circulating factors. that replacement contributes of action of to. and plasma

Indeed, there of a deficient may be the exchange on (TTP) ( 10). immune

is compelling plasma comprincipal mechIn thrombotic Finally, function many have

thrombocytopenic other theoretical

purpura effects

been
ment

proposed.
products,

These
fibrinogen,

include
and

depletion
possibly

of complecytokines,

myeloma.

lasma exchange refers to the removal of large quantities of plasma (usually 2 to 5 L) from a patient and replacement by either fresh-frozen or stored plasma. The procedure is frequently referred to as plasmapheresis when a solution other than plasma (e.g. , isotonic saline) is used as replacement fluid (apheresis from the Greek for to remove or to take
oped

improvement in reticuloendothelial system function ( 1 1 , 1 2) and immunomodulatory actions such as alterations In Idiotypic /anti-idiotypic antibody balance ( 13, 14). The influence of plasma exchange on cellular immune mechanisms, if any, are poorly defined.

TECHNICAL
The cussed

CONSIDERATIONS

away).
in the

Apheresis
1 950s to the

technology
harvest for transfusion techniques

was
peripheral

Initially blood

develcells

from ( 1 ,2).
widely, studies, array

healthy donors Subsequently,

into
were

patients introduced

usually based as primary or of human conditions

on anecdotal
adjunctive in which

or uncontrolled treatment for an circulating fac-

aspects of plasma exchange are disin most major textbooks of dialysis (e.g. , References 15. 16). The basic components of any modality of plasma exchange are withdrawal of yenous blood, separation of plasma from cellular components. and reinfusion of cellular elements plus either autologous plasma or an alternate replacement solution. Plasma Is separated from blood cells by either 1 ). At
devices:

technical extensively

tors were believed to contribute to disease pathophysiology. The 1 980s and 1 990s have witnessed a more rigorous re-examination of the efficacy of therapeutic plasma exchange (2-6).
1 Received Correspondence Hospital, 75 July 31
.

centrifugation present, there

intermittent

or membrane are two types

and continuous.

filtration (Figure of centrifugation

With intermit-

1995. Dr.

Accepted

October

27. 1995.
Renal Division, Brigham & Womens

to

Fran#{231}ois Madore,

Francis Street,

Boston,

MA 02115.

104&6673/0703-0367$03.00/0
Journal Copyright of the American C 1996 by the

Society of Nephrology American Society of Nephrology

tent devices, blood Is drawn in successive batches and separated into its components. These cycles are repeated as often as necessary to obtain the desired volume of plasma to be removed (usually, the equivalent of one plasma volume or 2.5 to 4.0 L is removed during a session). Advantages of intermittent devices are the simplicity of operation, the portability of the machines, and the possibility of a single-needle pe-

Journal

of the American

Society of Nephrology

367

Plasma Exchange

and Kidney Diseases

TABLE 1 Possible mechanisms exchange in kidney diseases

.

of action

of plasma

Removal

of an abnormal physiologic factor . Antibody (anti-GBM S Immune complexes

circulating factor or a produced in excess: disease) (lupus nephritls,

Patient

cryoglobulinemia) . Myeloma protein (myeloma cast nephropathy) . Toxic factor ? (1TP/HUS, FSG#{176})
Replacement of a deficient plasma . Thrombotic thrombocytopenic factor: purpura

Other effects on the immune system: . Removal of Inflammatory mediators . Improvement in reticuloendothelial
S Effects
a rrP/HUS. syndrome;

system

function
B

on Immune

regulation uremic

thrombotic thromboCytopenic purpura/hemolytic FSG. foca) segmental glomeruloscleros)s.

rlpheral vein). (>4 h large mL).

venous access (usually a large antecubital The disadvantages are the slowness of operation to perform a typical treatment) and the relatively extracorporeal blood volume required (>225 With the use of continuous-flow equipment, blood is fed continuously into a rapidly rotating bowl or channel In which red blood cells, leukocytes, platelets, and plasma separate into layers. Any layer or layers can be removed, and the remainder, together with a replacement fluid, can be returned to the patient (Figure 1A). Continuous-flow devices have the advantages of speed and automation. A microcom-

Figure 1 Centrifugal separator (A) and membrane filtration systems (B) for plasma exchange. (A) Blood is pumped into
.

puter lection

devices are relatively user friendly and less labor intensive than intermittent devices. The disadvantages are the high cost, the complexity, the relative Immobility of the equipment, and the requirement of two venipunctures (or insertion of a dual-lumen cath-

monitors most operations: anticoagulation, procedures, and fluid replacement. These

col-

the separator container. As the centrifuge revolves, different blood components are separated into discrete layers, which

can be harvested separately.

centrifuge into a collection and platelets are returned

Plasma is pumped

out of the

chamber. Red cells, leukocytes, to the donor, along with replace-

mentfluld.
brane cellular System System

(B) Blood is pumped

Into a biocompatible

mem-

eter). Membrane
centrifugal through continuous a

technology
devices. The parallel-plate

provides

an

alternative
Is pumped ifiter at of 50 usually to

to
a

patients blood or hollow-fiber

that allows the filtration of plasma while retaining elements. Suitable filters include Asahi Plasmaflo (Asahi Medical Co., Tokyo, Japan), Cobe Centry TPE (Cobe Renal Care Inc., Lakewood, CO), Toray Piasmax System (Toray Industries Inc., The Woodlands, TX), and Fresenius Plasmaflux (Fresenius AG, Oberursel, Germany).

mL/min pore diameters

passage

blood-flow (Figure 1B).

rate, typically The membranes

200 have device a plasma expected the average and

of 0.2
while

to 0.6

m,

sufficient
cellular

to allow
elements. weighed at the Infusion

the
removal with a time

patients
rate

hematocrit
of 30 to

value.

In general,
can be Thus,

of plasma

retaining

The
regular

separated
Intervals

plasma
(every

Is collected
15 to 30 mm)

and
and

blood-flow
required

rate

of

50 mL/min 100 mL/min.

rate

of the

replacement

fluid

is adjusted

manually
Intradevices, Equip-

or

brane

filtration

automatically to maintain vascular volume. Compared membrane filtration has

a stable circulating with centrifugal several advantages. relatively thus minimal; monitors are be performed

membrane the exposure

ment requirements blood pump and Membrane filtration

standard

are

pressure
can

hemodialysis-delivery

equipment

only a required. by using and pa-

face, such (although the clinical to be determined).

catheter is always

to perform a typical memthan 2 h. On the other hand, ifiters have potential problems related to of circulating blood to an artificial suras cell damage and complement activation

is less

significance of the latter has yet Also, insertion of a large vein

required to obtain adequate ifitration has bloodbeen

flow

rates.
membrane

tients
ysis tially,

with acute renal failure who require hemodialand plasmapheresis can receive both sequenusing the same dialysis machine. The procedure

In terms of efficacy, shown to be as safe mapheresis techniques

and

efficient

as centrifugal
difference is the

plas-

( 1 7). The
in economic

time

depends

on

the

blood-flow

rate

through

the

primary terms

between the cost of equip-

368

Volume

Number

1996

Madore

et a)

ment.
vice

tion,

with
standard

An automated continuous-flow centrifugal decosts approximately $40,000. Membrane ifitraon the other hand, requires only a blood pump pressure monitors, which Is available on every
hemodlalysis are roughly machine. Both techniques

general benefit

however, treatment

comparable In terms of cost per $200/treatment for nursing time, blood lines, and other basic supplies, plus $250 to $500/treatment for blood products). More sophisticated techniques achieve more selective removal of Ig and other molecules. These techniques Involve secondary treatment of separated plasma with procedures such as cooling (e.g. cryofiltratlon to remove cryoglobulins), perfusion through columns with adsorbents (e.g. immunoadsorption), or membranes with different pore diameters and filtration / adsorption characteristics (cascade filtration) ( 15). The treated plasma is then reinfused to the patient, thus minimizing loss of albumin and coagulatlon factors. These techniques are not routinely (approximately
, ,

care over time may be misconstrued as a of plasma exchange. Second, the natural history of many diseases commonly treated by plasma exchange (e.g. cryoglobulinemla, systemic lupus erythematosus [SLE]) Is characterized by episodes of exacerbation and remission, further underscoring the
,

importance

of

adequate

concurrent

controls.

Third,

the
ment dering

threshold
protocols It difficult

for intervention

and

may vary widely to compare studies.

the details of treatbetween centers, renFourth, plasma

exchange Is primarily utilized in the treatment of inflammatory renal diseases as an adjunct to conventional Immunosuppressive therapy and might be cxpected a priori to confer only a small additional benefit

that
Finally,

would

require

large
studies

sample
are

size

for its detection.

less likely to be

negative

inevitably

published and estimations basis of published reports plasma exchange.

of efficacy made on the may be biased in favor of

available

In most

centers

and

are

usually

reserved

for

research purposes. The typical replacement fluids are fresh-frozen or stored plasma, 5% albumin, or other plasma derivatives (e.g. purified protein fraction), and crystallolds (e.g. 0.9% saline, Ringers lactate solution) ( 15). The choice of replacement fluid has major Implications for the efficacy of the procedure, oncotic pressure, coagulatlon, and potential side-effects. Albumin or purlfled protein fraction Is usually preferred to plasma, given the potential risks of hypersensitivity reactions and transmission of viral infections with the latter. In
, ,

EFFICACY OF PLASMA RENAL DISEASES

The efficacy most extensively
rapidly

EXCHANGE

IN SPECIFIC
has been secondary assessed forms

of plasma exchange in primary and

glomerulonephritis

of
In-

progressive

(RPGN),

cluding anti-glomerular basement membrane (antiGBM) antibody disease, immune complex-mediated glomerulonephritis, and pauci-immune RPGN. Other disorders In which the technique Is often advocated Include TTP, acute renal failure associated with myeloma, and allograft rejection.

this

regard,

albumin

(5%)

Is generally

combined

with

0.9% salIne composition to the needs

example, high risk

on a 50%/50% of replacement of individual

may be
,

(vol/vol) fluids patients

more

basis. The precise should be tailored and diseases. For

In patients at

Anti-Glomerular Basement Antibody Nephritis

RPGN

Membrane

plasma ofbleeding

prudent

inated
syndrome apy (e.g.

intravascular
,

IHUSI/TFP, daily exchanges

(e.g. severe liver disease, dissemcoagulation, hemolytic uremic etc.) or requiring intensive therfor weeks) (6).

Antl-GBM antibody disease typically presents as without or with pulmonary hemorrhage (Goodpastures syndrome). Renal biopsy characteristically reveals crescentic glomerulonephrltls with deposition of immunoglobulln G (IgG) and C3 along the glomerular basement membrane ( 18). Greater than 90% of
patients have anti-GBM antibodies In their circula-

INTERPRETATION OF CLINICAL TRIALS EVALUATING THE EFFICACY OF PLASMA EXCHANGE: GENERAL COMMENTS
Before examining the published studies on the efficacy of plasma exchange In renal diseases, several general considerations that may complicate data Interpretation deserve mention. First, there are few prospective controlled clinical trIals of adequate statistical power to allow definitive conclusions to be reached regarding the therapeutic value of plasma
exchange.
rarity of

tion. The latter are directed against the 28-kd noncollagenous C-terminus of the a3 chain of the Type IV collagen, an epitope that Is relatively restricted to glomerular and alveolar basement membrane. In general, disease activity correlates with the titer of circulating antibodies, and passive transfer experiments have provided compelling evidence that circulating
antl-GBM servatlon peutic antibodies provided removal of are nephrotoxlc. The latter oban attractive rationale for theraanti-GBM antibodies by plasma

This
most of

drawback
the disorders

reflects,
under

in part,

the

relative
To

Investigation.

exchange. Before 1975, antl-GBM-induced dated with very poor prognosis.

nephrltls In a study
patients with

was assoby Wilson

antl-GBM

compensate,
grouped

many
heterogeneous

Investigators
diseases,

have

understandably
often retrospec-

and

Dixon

in

1973,

89%

of

tively, and utilized sign is potentially nosis, recognition

historical hazardous, of milder

controls. The latter degiven that earlier diagcases, and Improved

nephrltis nations

who had been treated with various combiof steroids and cytotoxic drugs progressed to ESRD or death within 5 yr of diagnosis ( 19). Similar conclusions were reported in a review by Rees, with

Journal

of the

American

Society

of Nephrology

369

Plasma

Exchange

and

Kidney

Diseases

more than 85% of patients treated with conventional therapy progressing to ESRD (20). Against this background, only two controlled studies evaluated the efficacy of plasma exchange as an adjunct to conventional immunosuppressive therapy in anti-GBM ne-

patients

that

received

plasma

exchange

had

milder

disease Johnson on the

phrltls

(2 1 ,22). Although these studies were small (17 and 20 patients, respectively), both suggested a beneficlal effect, as evidenced by more rapid decline in anti-GBM antibody titers, lower mean serum creatinine values at end of therapy, and fewer patients progressing to renal failure (Table 2). Johnson et at. compared the influence of plasma exchange (4 L every

than patients in the control groups. Indeed, et at. reported that the percent of crescents Initial biopsy and the Initial serum creatinine values correlated better with outcome than did the therapeutic modalities employed. The results of more than 20 uncontrolled studies on almost 250 patients, published over the past 20 yr.
also suggest a favorable effect

about
largest

40%

of patients
series

(reviewed
comes

on renal outcome In Reference 23). from the Hammer-

in The

published

3 days)

plus

immunosuppression

(prednisone

and

cyclophosphamide) to Immunosuppression alone on the clinical course and rate of disappearance of antibodies In a randomized controlled trial of 1 7 patients with antl-GBM disease (2 1 ). Patients treated with plasma exchange had a more rapid disappearance of antl-GBM antibodies and a mean serum creatinine value that was less than 50% that ofpatients receiving immunosuppression alone at the end ofthe study (9.5 0.7 mg/dL 1840 62 pmol/Ll versus 4.4 0.6 mg/dL [390 53 mol/Ll; P < 0.05). In a nonrandomIzed study, Simpson et a!. compared the clinical course and levels of anti-GBM antibodies in eight
patients exchange treated (at least

smith Hospital, a pioneer in this field. More than 59 patients have been treated with plasma exchange and immunosuppression at this center since 1974 (20,2429). When using daily 4-L plasma exchanges for at
least 14 days, 85% of patients were alive after 2

months of follow-up, 41% progressed to ESRD and required chronic dialysis, and 44% retained Independent renal function. These results compare favorably with historical controls used before 1975 (survival,
47%; ESRD, 87%). Despite plasma exchange, significant recovery of renal function was Infrequent In patients who were either oligurlc, had a serum creatinine value above 600 mol/L (6.8 mg/dL), or required dialysis at presentation, even though antl-GBM anti-

with

immunosuppression 10 sessions, 3 L each),

and
four

plasma patients

treated with immunosuppresslon alone, and eight patients who did not receive specific therapy (22). There was a more rapid fall in the level of ant!-GBM
antibodies in the eight patients treated with immuno-

body titers were reduced effectively. In light of these studies, It seems reasonable to conclude that plasma exchange, when used as an adjunct to conventional Immunosuppressive drugs,
accelerates from the Instituted the disappearance of anti-GBM circulation and may Improve renal promptly In patients with milder antibody function forms if of

suppression and plasma exchange. In this group, only three patients progressed to ESRD, compared with two of four patients treated with immunosuppresslon alone and six of eight patients with no specific therapy. It should be noted, however, that at study entry,

anti-GBM (oligurla, mol/L

ably be

nephritis. In patients with dialysis, or serum creatinine [6.8 mg/dLl), plasma exchange
reserved for treatment of lung

severe disease value > 600 should probhemorrhage

TABLE 2. Major clinical trials evaluating the efficacy basement membrane antibody nephrltisa

of plasma

exchange

in treatment

of anti-glomerular

Benefit of Plas ma Exchange

Authors
Patients

on

Number

of

(Reference)

Study

Design

Dialysis at Presentation

Plasma Exchanges

Concomitant

Renal

Therapyb

Outcomec
PE Controls
67%d

Mortality

PE

Controls

Johnson
(21)

et ci. et a!.

Randomized
controlled Nonrandomized trial

17
20

2 of 17
7 of 20

4 to 17
10 to 25

Steroids
Cyclo Steroids

25%d

25%
13%

11%
25%

Simpson (22)

38%d

50%d

controlled
Uncontrolled

study
case 59 30 of 59 >14

Cyclo
Aza

Hammersmlth

Steroids

41%f

87%

15%f

47%f

Groupe (20)
0

series
ESPD. end-stage

Cyclo
Aza
renal disease.

b
C

PE. plasma exchange; Cyclo. cyclophosphamide; Aza. azothioprine; All concomitant methods of therapy were administered by mouth.
Expressed as incidence of ESRD. Significant benefit. but patients receiving plasma From Hammersmlth Hospital. London. England-initial exchange had experience

d e

milder disease. from 1975 to 1985 reported

in references

(24-29).

I Benefit

when

patients

treated

with plasma

exchange

are compared

with historical

controls.

370

Volume

Number

1996

Madore

et al

because renal function is unlikely to recover even with aggressive treatment. As an initial regimen, most authorities recommend daily plasma exchange for 14 days, using 4-L exchanges and albumin solution as replacement fluid, in addition to steroids and cytotoxic drugs. Response to therapy should be monitored by repeated assessments of urine output, serum creatinine values, and plasma antl-GBM levels.

with

removal

of 1 plasma

volume

at each

session.

At 1,

Pauci-Immune Glomerulonephritis

Rapidly

Progressive

Approximately 40% of patients with RPGN present with crescentic glomerulonephrltls characterized by few or absent immune deposits (pauci-immune RPGN), as determined by direct immunofluorescence (23). Patients with this clinicopathologic presentation
usually have either Wegeners granulomatosis, poly-

3, 6, or 12 months following randomization, there was no significant difference between the two groups In mean serum creatinine values, change in serum creatinine value, or with regard to dialysis dependency. Although the aforementioned two studies suggested that routine plasma exchange affords little benefit In the majority of patients who present with pauci-Immune RPGN, Pusey et at. (35) provided evidence for a benefit in a subgroup of patients who presented with severe (dialysis-dependent) disease. Specifically, they compared the influence of plasma exchange plus immunosuppresslon (steroids, azathioprine, and cyclophosphamide) to Immunosuppresslon alone on renal

arteritis nodosa, or renal-limited pauci-immune GN. These diagnoses may represent a spectrum of manifestatlons of a single disease because there is marked overlap of clinical and histopathologic features, and many patients have anti-neutrophil cytoplasmic antibodies (ANCA) in their circulation. In some, although not all, studies, circulating ANCA titers correlate with disease activity, and ANCA may contribute to the pathophysiology of pauci-immune RPGN through reactivity with neutrophils and/or endothelial cells, has been and poor.

outcome in 48 patients with paucl-immune RPGN. The plasma exchange regimen consisted ofat least five 4-L exchanges within the first week. The total number of exchanges was then determined by the clinical response. A mean of nine procedures was performed (range, 5 to 25). Of the 1 1 dialysIs-dependent patients who were treated with plasma exchange, ten recovered sufficient renal function to stop dialysis (mean follow-up of4 months), compared with only three of eight conventionally treated patients (P = 0.041).
A similar trend was noted in two other trials. Rifle

other Precise
ture, several

inflammatory figures
because types

mechanisms
RPGN

(30-32).
in general

The prognofrom the literawith would

and Dechelette randomized 14 patIents with 1dbpathic RPGN, many pauci-Immune, to receive pulse methylprednlsolone, immunosuppressive drugs and heparmn with (six patients) or without (eight patients)
,

515 of pauci-Immune

plasma

exchange

(36).

Recovery

of

renal

function,

are

difficult

to obtain

most series consist of of RPGN. However, available

patients data

indicate
without

that
therapy

80%

of such
with high

patients
dose

progress
immunosuppression

to ESRD Refer-

or cytotoxic drugs (see the ence 23). Five randomized controlled efficacy of plasma exchange tional Immunosuppressive
pauci-immune randomized 26 RPGN (Table patIents with

review

by

Couser,

measured as discontinuation of dialysis or decrease of at least 50% in serum creatinine value, was better in the plasma exchange group at 2 months after randomizatbon (P = 0.02); however, this difference was no longer significant at the end of follow-up (22 months). Maurl et at. prospectively compared the efficacy of an immunosuppressive regimen of steroids and cyclophosphamide with or without plasma exchange in a

trials have evaluated the as an adjunct to conventherapy in patients with 3) (33-37). Gl#{246}ckner et at.
RPGN to immunosup-

pressive

prine) ( 14 and 12 patients, respectively) (33). The plasma exchange regimen consisted of 3.5-L exchanges performed over 4

agents (steroids, with or without

cyclophosphamide, plasma exchange

azathio-

wk:
sessions

three

sessions
weekly
was

in the
thereafter.
found in

first
No
renal

week

and
outcome

at least
significant
between

two

statistically

difference

groups. After 8 wk, 73% of patients tional therapy and 69% undergoing exchange showed an Improvement
creatinine clearance values compared

receiving convenadjunctive plasma of > 20 mL/min in

with pretreat-

heterogeneous group of 22 patients with RPGN, most of whom had paucl-immune disease (37). Among the 1 1 patients who presented with a serum creatinine value above 800 mol/L, renal function improved in five of six patients treated with adjunctive plasma exchange, compared with only one of five patients treated with immunosuppressive drugs alone (at 4 months after randomization, serum creatinine value was 728 tmol/L with plasma exchange versus 1163 mol/L in control patients; P = 0.016). Importanfly, none of the randomized controlled trials listed in Table 3 reported improvement in patient survival when this
parameter was specifically addressed.

In aggregate, argue against

forms

the results of five randomized a role for plasma exchange in

RPGN, but suggest

trials milder

ment values, or attained a clearance value of greater than 10 mL/min (for patients who required dialysis at presentation). Cole et at. randomly assigned 32 patients with renal-limited pressive therapy (steroids RPGN to receive immunosupand azathioprine) with or

ofpaucl-lmmune

a potential

benefit when the technique Is used as an adjunct to conventional immunosuppressive therapy in patients with severe disease. This relative lack of efficacy probably
serving

reflects
renal

the

without plasma exchange (34). Patients underwent

( 1 6 patIents In each group) at least 10 plasma exchanges,

suppressive

agents

efficiency of in halting
in most

conventional inflammation
patients.

immunoand preThe latter

function

Journal

of the

American

Society

of Nephrology

371

Plasma

Exchange

and

Kidney

Diseases

U)

Q 0
V

. Cv)

.9
0
.4=

.Q

ci)
>

ci) 0) 0 0. 0 0. 0 ci)
C

0 0) C 0 C C)

.t0 0

J?

2
C u0

22
w0tLJ#{176}

fi i
0.

oa.O

oa.O

oa.O

x
0

LU

E
Cl)

0 Q. 0

a)

E
0
0

T :- :
00
-

ilJ
0 0..
.

a
.-

a)
C a)

0
0 C a)

r.. Q .. ?;=0.

,00
CV) --;;

#{176} 0.. ioE

i!
E0
.C,)

#{149}:
0 .c:

#{176}2
.. =0.Cc ..

0 v-

E 5
LC

s,,, .i2
C

E E
C)

2ooo,0)vvo . cQ
-

gci.O

a)vVo0)0.0Eo cLO
LL.

5
Z
.

8
,

C 0 0.

Q.Q
a) 0 C

a
0

0 0.
C

C
,

E
0 a)a)

0.0

-B

,Bo ti

.c

E
.

a,B2,;2
5o
,-% Cl)

a.2O:r

22
5L)

ci)

g
b
a
0 0

E
0
C

0
-

e
0 C

a)

C 0

a)

F-

a 0
S-.

2
D

a)EC

,, (I)

0 C)
C

E
:,:

g
0
-

#{176}C 0

,.:
,-

%-,

.?

0 0 ti) 0
..

0
C

0 x
ci)

?: P
0

z
+
.

z
.

E
U)

%., .

0 0. 0
0
CC

5) 0

0 C

a) a) 0.

0 C

a) a) 0.

0 C

a) a) 0.

0 C

a)
-

a)
C

0 0
C)

a) 0.

. a)

. a)

. a)

:a)
Cl)

Cl)

Cl)

Cl)

ci) ci)
C

.2
0 Q 0
,

.2
0

.2
V 0 c) 0 a) 0 0

2
V

to

aQ
C)

Q
O 0 a) 0 0 a) 0

0) C 0
C

0 a) 0
-

a)

0 > ci)
Cl)

0
0

o3

S2

.2

C
Cl)

Ewe)
a)

E
0CC)

E
QCl)

EQCI)

0 0 ci) 0
C
0

a)

a)

a)
.0

.a)

a)

z t5

O ci 0 N.

C4 C)

U)

, ,-

c1

0
.

00

.A

.2

E#{246}
0

2. E#{246} E#{246}
0 0

V0

2. E
0#{176}

0-_

Va

O.c
Cci)
-. . ....

a)

Sc,
DC a)
<
-

00

Ct)
: >.

CC
0

ii:
372

=R

:28 (

#{246}2 U

i2
Q.

,ar2,

Volume

Number

1996

Madore

et al

conclusions results of

are 12

further uncontrolled

supported case

by series

the

combined in

pmol/L

( 1 .2 mg/dL)

and

remission

of

24-h

urinary

(reviewed

Reference

23),

suggesting

a response

rate

of 70%

in

patients with RPGN treated with plasma exchange similar to that of patients treated with conventional therapy (response rate, 60%). Given the paucity of convincing data, It is impossible to give firm recommendations regarding the specifics of therapy. If uti-

lized, It would seem prudent plasma exchange sessions

immunosuppressive and albumin solution output, titers.

to perform during the

at least four first week of

4-L fluid. exchanges Response

protein excretion to 0.2 g per day (P = 0.86). SIx patients ( 13%) in the standard therapy group and eight patients (20%) in the plasma exchange group died (P = 0.39). Indeed, there was a tendency for patients treated with plasma exchange to have a worse outcome, which ultimately led the External Data Monitoring Board to recommend early termination of the trial. Four other randomized controlled trials of plasma

therapy, using as replacement

exchange
the and last

in lupus
15 yr (Table

nephrltis
4) (47-50).

have
Web

been reported over et at. randomized

to therapy
ments possibly

should
ANCA

be monitored
serum

with
creatinine

repeated
values,

assess-

of urine

Lupus

Nephritis

Clinically overt nephritis complicates 38 to 90% of cases of SLE. Clinical manifestations of lupus nephritis range from mild abnormalities of the urinary sediment to fulminant Inflammation and renal failure. These diverse clinical presentations, in turn, reflect the array of renal histopathologic entities encountered in SLE, classified by the World Health Organization as follows: Class I, normal or minimal disease; Class II, mesangial injury; Class III, focal and segmental proliferative GN; Class IV, diffuse proliferative GN; and Class V. membranous GN (reviewed in Reference 18). Typical serologic abnormalities in lupus nephritis in-

20 patients to receive either or sham procedures over patients with mild disease clearance value > 20 mL/min, mg/kg, no cytotoxic drugs). nonsteroidal anti-inflammatory roids were maintained at

exchanges (47). Only (creatinine prednisone dose < 1.0 Throughout the study, all drugs and corticosteconstant dosage. Plasma exchange produced significant reduction In circulating immune complexes and anti-DNA antibodies, but the frequency and degree of partial or complete remission was the same in both plasma exchange and control groups. Clinical outcome was evaluated by using physician assessment and a clinical activity

six 4-L plasma a 2-wk period were included

index

designed

on

the

basis

of 2 1 separate

clinical

dude

the

presence

of anti-nuclear

(ANA),

antl-dou-

bled-stranded DNA (anti-dsDNA) and other autoantibodies, circulating Immune complexes, and evidence of complement activation. Formation of Immune com-

manifestations of active SLE. Among the 18 patients who completed the trial, the proportion of patients with at least 50% improvement in physician assessment and clinical activity Index was 55% in the plasma exchange group versus 33% in the control group at 6 wk after entry into the study, a difference that did not reach statistical significance. Derksen et

at.

conducted

a trial

that

compared
cytotoxic had not

a 3-wk
drugs responded

course
(cyclophos-

of

plexes within glomeruli appears to be a central event in the pathophysiology of lupus nephritis. Plasma exchange has been advocated as an adjunct to conventional immunosuppressive regimens to remove immune complexes, auto-antibodies, and other inflammatory mediators, and to improve reticuloendothelial function (11,12,38-40). Early case reports and uncontrolled case series suggested a benefit of plasma exchange in lupus nephrltis (40-45), however, a recent large multicenter

plasma exchange versus phamide or azathioprine) ative lupus nephrltis who

in 20 patients (

with

proliferwithin 3

wk

to high

doses

of prednisone

1 mg/kg)

(48).

Five

of 20 patients had a significant clearance values (three in the

increase PE group;

in creatinine two in the

cytotoxic drugs group). Of the six patients with a creatinine clearance below 20 mL/mln (two in the PE group, four in the cytotoxic drugs group), none had a significant change In clearance. No beneficial effect of plasma exchange on extrarenal manifestations was observed. The French Cooperative Group randomized five patients to corticosteroids and plasma exchange and seven patients to corticosterolds alone. All patients received 1 .5 mg/ kg per day of prednisone for 60 days, with subsequent tapering to reach 1 mg/kg per day at Day 90. The plasma exchange regimen consisted of 23 sessions over a 2-month period, with removal of at least 60 mL of plasma / kg per exchange. Clinical outcome was evaluated by using physician assessment and a clinical activity index obtained at Day 90. Outcome was similar in both treatment groups (activity scores: 34 In the PE group, 39 In the control group; P > 0.05). In contrast to these studies with short-term intensive therapy, Clark et at. suggested a possible role for

prospective
strong evidence

randomized
against its

controlled
use (46).

trial
The

provided
Lupus Ne-

phritis

Collaborative

Study

Group

assessed

the

effi-

cacy of plasma exchange as an adjunct to prednisone and cyclophosphamide in 86 patients with severe lupus nephritis (mean duration of disease > 13 months; serum creatinine value > 2.0 mg/dL I 180 mol/Ll). Patients underwent plasma exchange three times weekly for 4 wk and were then followed for an

average of 136 rapid reduction

wk. Plasma exchange of serum anti-dsDNA

induced a more and cryoglobu-

lin, but did not influence renal function or mortality. A similar percentage of patients (28% in the standard therapy group versus 30% in the plasma exchange group) underwent remission of their renal disease, as judged by a return of serum creatinine to 106

Journal

of the

American

Society

of Nephrology

373

Plasma

Exchange

and

Kidney

Diseases

TABLE 4. Randomized lupus nephritisc

Authors

controlled

trials evaluating

the efficacy

of plasma

exchange

in treatment

of

Number (Ref.)
Lewis et a!. (46) Wel et a!.
(47)

Stud

Desi

Disease

Initial

Renal

of

Concomitant

Benefit of Plasma Exchange

Clinical Outcome Mortality
Mortaiity:c

Severity
Severe Mean

Function
s. creaf. of

Plasma
Exchanges
12

Therapyb
Steroids ESRD:c

Randomized

86

controlled
Randomized controlled

trIal 20
trial

disease Mild disease

180 moI/L
CrCI >20 mI/mm 6

Cyclo
SteroIds AntI-malarlals

PE: 25% Controls:

Activity index

17%
(>50%

PE: 20% Controls:

NP

13%

lmprovement):c

NSAID
Derksen
(48)

PE: 55%
Controls: 33% NP

et a!.

Randomized

20 trial

controlled

Unresponsive to conventional therapy

Mean CrCI of 30 mL/mln

Steroids

Remission rate:c PE: 33% Controls: 18% Activity score:c

PE: 34

French Group

(49)

Randomized controlled Randomized controlled

trial

12 ActIve disease 39 Mild disease

Patients with <50%

crescents#{176}

23

SteroIds

NP

Controls: 39 Clark et a!. (50)

CrCI >30

mI/mm

NR

trial

SteroIds Aza

S. creatinlne:c PE: 97 ,mol/L Controls: 124 ,moI/L Aza. azathiopnne; Cyclo.

Mortallty:c

PE: 5% Controls: 0% cyclophosphamide;

PE. plasma exchange; s. creat. . serum creatinine; CrCI. creafinine clearance; NSAID. nonsteroldal anli-infiammafory drug; ESPD. end-stage renal disease.
All concomitant methods of therapy were administered by mouth.

NP. not reported; group.

were

b
C

d
Cl

No statistically significant difference between plasma exchange group French Cooperafive Study Group on Systemic Lupus Erythematosus.
Patients wIth rapidly progressive glomerulonephrlfis or >50% crescents

and control
on renal biopsy

excluded.

prolonged

plasma

exchange
study, randomized

therapy
39

in lupus

nephri-

Cryoglobulinemia
Cryoglobulins are Ig that precipitate in the cold.

tis (50). In a controlled proliferative GN were

munosupressive with or without

A trend toward

therapy 4-L plasma

better

patients with diffuse to conventional im(steroids azathioprine)

exchange

every
of renal

3 to 4 wk.
function

preservation

was identified In the plasma exchange ically, the mean serum creatinine lower in patients treated with plasma
the control group
.

group. Specifvalue was 33% exchange versus

a difference

at the

end

of the

study,

that did not reach exchange, 1 1 mg/dL apy, 1.4 mg/dL 1124 In summary, the
randomized controlled

statistical significance (plasma [97 mol/L], conventional thermol/Ll, P = 0.085) results of multiple prospective
trials do not support a role for

plasma

exchange in the routine treatment of lupus nephritis. There is experimental and clinical evidence that rapid removal of circulating antibody by plasma exchange triggers a rebound B-cell clonal proliferation and enhanced antibody synthesis (13, 14, 5 1 ,52). Because proliferating cells have increased vulnerability to cytotoxic agents, It has been suggested that plasma exchange may be useful in patients with lupus nephritis if synchronized with pulse cyclophosphamide (the

latter administered shortly after plasma exchange) (53). Initial studies supported this contention (54-56); however, this approach requires validation in larger trials before It is employed as routine therapy.

Three types of cryoglobulinemia have been described (57,58). In Type I, the cryoglobulin is a single monoclonal Ig and is usually found in association with multiple myeloma, Waldenstr#{228}ms macroglobulinemia, or other lymphoproliferative malignancies. Types II and III are mixed cryoglobulins that contain at least two 1g. Type II cryoglobulins are usually composed of polyclonal IgG directed against an antigen and a monoclonal 1gM with rheumatoid factor activity directed against the polyclonal IgG. In Type III cryoglobulins, both Ig components are polyclonal. Most mixed cryoglobulins are found in association with connective tissue diseases, infections, lymphoproliferative malignancies or autoimmune disorders and are therefore referred to as secondary mixed cryoglobulinemlas. Until recently, no precise etiology could be found in approximately 30% of cases of mixed cryoglobulins and these were termed essential mixed cryoglobulinemla (58). Recent studies suggest that most cases of mixed essential cryoglobulinemia cases are triggered by hepatitis C (HCV) Infection (58). Precipitation of cryoglobulins within the glomerular capillary lumen can induce a spectrum of nephritides, depending on the rate and magnitude of deposition. Renal symptoms usually appear several years after

374

Volume

Number

1996

Madore

et

al

the

onset

of the

extrarenal

manifestations.

Fifty

per-

nonsteroidal
cytotoxic mortality drugs rate

anti-inflammatory
(6 1 ). Gorevic at 7.4 yr In

drugs,
et at.
patients plasma
,

steroids,
reported
who had

and
a 55%
been

cent of cases present with proteinuria, microscopic hematuria, or mild to moderate renal insufficiency. Another 25% of cases present with nephrotic syndrome, whereas 20 to 25% present with nephritic
syndrome. The characteristic morphologic glomerular

treated
Considering

with

conventional
these poor

Immunosuppression
results, exchange

(60).
was

proposed

as a means

of rapidly

reducing

the

circulatIn

lesion is membranoproliferative GN with subendothehal deposits (58). GN Is triggered by cryoglobulininduced complement activation, leukocyte infiltration, and induction of other mediator systems. Regarding relevance to plasma exchange therapy, most investigators report poor and disease activity tial cryoglobulinemia correlation between (59,60). In addition, has a variable and the mixed often cryocrit essenunpre-

Ing cryocrit and addition, removal

limiting deposition of large quantities

in tissues (62). of cryoglobulins

could

potentially

restore

the

clearing

function

of an

dictable
renal progress

natural history. In one third disease has an indolent course

to renal failure despite

of patients, and does

the not

overloaded reticuloendothelial system ( 1 1). Plasma exchange has been utilized for the treatment ofcryoglobulinemia for over 20 yr. unfortunately without being subjected to prospective randomized controlled clinical trials. Table 5 summarizes the results of uncontrolled studies in which greater than five

persistent

urinary

abnormalities. Other patients have either a waxing and waning course characterized by exacerbations and remissions, or slowly progressive renal injury that culminates in end-stage renal failure (58). Progression to ESRD was initially reported in 30 to 50% of patients who had been treated with various combinations of

patients ported severe

with corticosteroids (methylprednisolone, (63). Five ofnine patients had significant in renal function and/or protelnuria 54 plasma exchanges over a 10-month

were evaluated (59,63-68). Fern et at. renine patients with mixed cryoglobulinemia and membranoproliferative GN who were treated plasma exchange alone or In combinatIon with
40 to 60 mg/kg) Improvement after a mean of interval (from a

TABLE 5. Larger
Authors

studies

assessing

value

of plasma

exchange

in treatment Number

of cryoglobulinemiaab of Concomitant
Therapy (method) Steroids (p0)

Patients with
.)

Response
Renal Functiond Improved s. creat. in 55% Improved

to Therapy
Mortalityc NR

( e

Study

Design

Renal Involvement 9 of 9 Mixed

Diagnosisc

Plasma Exchanges 15 to 1 13

Fern ef a!. (63) Singer (64)

Uncontrolled case series Uncontrolled

cryo

et a!.

16

10 of 16

Mixed

cryo

3 to 12

Steroids

(p0)

25%

case series

(mostly

Type II)

Cyclo

(p0)

or stable CrCI in
80% Improved 5%

Sinico et a!.
(65)

Uncontrolled

20

16 of 20

Mixed

cryo

3 to 34

Steroids

case series
et
Uncontrolled 15 8 of 15 Mixed cryo 3 to 5

(iv+po)
Cyclo (po) Steroids (p0) Aza (p0)

s. creat.
in 87% Improved

Valbonesi
a!. (66)

13% in 62%

case series

clinical
score

75%
Frankel (59)

et a!.

Uncontrolled case series

13

10 of 13

Mixed cryo (Type II)

4 to 238

Steroids (p0) Cyclo (p0)

Improved or stable
5. creat.

LAbbate et a!. (67)

Uncontrolled

11

1 1 of 1 1

Mixed

cryo

4 to 28

Steroids

in 67% Improved

9%

case series

(iv+po)
Cyclo (p0) Arabinoside-C Steroids (iv + p0) Aza (p0)
creatinine;

or stable
CrCl in 73% Clinical Improvement in 75%f crcl. creatinine

Schena

et a!. Uncontrolled

10

7 of 10

Mixed

cryo

12 to 17

NR

(68)

case series

a
b C

cryo. cryoglobulinemia; Aza. azathioprirte; Cyclo. cyclophosphamide; NR. not reported; s. creat. . serum Includes studies published before 1995 that involved more than 5 patients. Most studies focused on mixed essential cryoglobulinemia and excluded cases with secondary disease.

clearance.

e f

Expressed Expressed
Estimated

as percent as percent
value (not

of patients of patients
precisely

with renal involvement from the entire group.

at presentation.

reported).

Journal

of the

American

Society

of Nephrology

375

Plasma

Exchange

and

Kidney

Diseases

pretreatment
mg/dL

serum creatinine value of 4.6 2.1 1406 185 mol/Ll to a posttreatment value of 2. 1 1 .0 mg/dL I 185 88 pmol/L]). Improvement was almost exclusively observed In those patients who had presented with a rapid worsening of renal function in the weeks before plasma exchange and/or in the
mal

cryoglobulinemia with the realization

is undergoing constant reevaluation that hepatitis C Is the etiologic

agent
quently

in

most
induces

cases
remission

and

that
(58).

interferon-alpha

fre-

IgA Nephropathy
IgA present nephropathy is hematuria the and most common form of GN up

presence
renal

of histologically
scarring. Singer

ported

on

1 6 patients

who

active lesions and miniet at. retrospectively rewere treated with either

worldwide
with

(69).

Whereas

the
normal

majority
renal

of

patients

function,

immunosuppressive therapy alone (three patients), plasma exchange plus immunosuppression (ten patients), or plasma exchange alone (three patients) (64). Ten of the 1 6 patients had proteinuria, with a range of

to 50% of patients suffer progressive GFR over 20 yr (69,70). Furthermore,

of patients with phritic syndrome almost identical the typical renal IgA nephropathy or RPGN. The morphologically presentation

deterioration in as many as 10%

neis to

0.2 to 12.8 g/day, these ten patients.

rienced renal a fall function

and the GFR was Fourteen of the

or

reduced in nine of 16 patients expelevels and In eight of ten

present with latter presentation and functionally of Henoch-Sch#{246}nlein as

In circulating stabilized

cryoglobulin improved

patients
the nor

who

had

renal

impairment
degree was

over

the

course

of

study; however, neither the the duration of the follow-up

of improvement reported. Sinico

purpura, and many investigators consider Sch#{246}nlein purpura and IgA nephropathy trum of presentations of the same pathogenic The pathognomonic feature of both diseases
biopsy is mesangial deposition of IgA. Serum

Henocha spec-

process. on renal
total IgA

et at. reviewed the clinical and laboratory data of 20 patients with mixed cryoglobulinemia who were treated with immunosuppressive drugs and an average of 18 plasma exchanges (65). Sixteen patients had renal involvement. Serum creatinine values and proteinurla decreased significantly with therapy in all but two patients (serum creatinine value, from 2.9 to 1.6
mg/dL [256 1 .6 g/24-h). to 141 Similar j.mol/Ll; results proteinurla, were reported from 3.5 to in several

concentration is elevated in 33 to 55% of adults with IgA nephropathy (71), although circulating levels of IgA do not usually correlate with the severity or activity of the disease. The precise mechanism(s) of glomerular IgA deposition and role of deposited IgA in disease pathophysbology have not been defined. Nevertheless, plasma exchange has been advocated as an adjunct to immunosuppressive drugs in patients with aggressive disease to remove circulating antibodies and other putative inflammatory mediators, and to augment reticuloendothelial system function (72). The published worldwide experience of therapeutic plasma exchange in IgA nephropathy or HenochSchOnlein purpura is limited to approximately 50 patients, all reported as case reports or small uncontrolled series (72-79). NIcholls et at. evaluated the efficacy of plasma exchange in 1 3 patients who had progressive IgA nephropathy (79). The plasma exchange regimen consisted of daily 3 to 4-L treatment for 4 days, followed by thrice-weekly treatment for 2

other series (Table 5) (59,66-68). Overall, plasma exchange has been reported to lower serum cryoglobulin levels and is associated with improved renal function in 55 to 87% of patients. It is claimed to be most effective if used in patients with active inflammatbon Plasma survival and in the first few exchange is also (an approximate weeks of an acute flare-up. associated with Improved 25% mortality rate) when

compared
mortality

with historical data (an approximate 55% rate). Because of the uncontrolled nature of all of the studies reported to date, It is impossible to determine If the improvement In renal outcome and in survival is
attributable to plasma exchange or to other factors such as patient selection, earlier diagnosis, wide vanability In presentation and natural history, advances in general medical management or other aspects of

wk, twice-weekly treatment until

plasma exchange.

the

treatment completion
Comparison

for

2 wk, then of 3 months

of the rate

weekly of total

of deterio-

ration in renal function plasma exchange suggested fect. In seven of 13 patients,

was efit significantly slower

before, during, and after a possible beneficial efthe rate ofdecline In GFR

immunosuppressive therapy. It Is probably advisable to limit plasma exchange to patients with clinical or renal biopsy evidence of acute active and severe disease. If employed, most investigators advocate cornbining plasma suppression weekly for the exchange with and performing first 2 wk, using conventional the procedure 4-L exchanges immunothrice and

exchange
was

than
transient,

before
as

(P

during treatment with plasma < 0.00 1 ). However, the benthe rate of decline returned to

pretreatment values after cessation of plasma exchange. There was also a suggestion that plasma exchange was more likely to be beneficial in patients with a rapidly progressive course. Coppo et at. studied
the effect of nisolone and five patients plasma exchange combined cyclophosphamide on renal with IgA nephropathy (74). with predfunction in Two of the

albumin solution as replacement should be tailored subsequenfly, clinical and biochemical response. circulating cryoglobulins is a poor because, as mentioned above, there
tion between the exchange cryocrit in the and disease therapy of plasma

fluid. Therapy according to the Measurement of index of efficacy is a poor correlaactivity. of mixed The essential role

three whereas
slowly

patients two
evolving

with other
course

a rapidly patients
had no

progressive
after plasma

course
exchange,

had

substantial

Improvement

presenting
substantial

with

a more

benefit.

376

Volume

Number

1996

Madore

et al

Despite

these

claims

of efficacy

in patients

with

actuarial
exchange

patient
as 77%

survival
with

was
the
=

observed
other

with
therapies

plasma
(100%

marked variability in the plasma exchange regimens. Given the encouraging preliminary results, however, the role of plasma exchange deserves to be assessed further in IgA nephropathy in larger randomized controlled tnals.

rapidly progressive course, no firm conclusion drawn regarding the role of plasma exchange nephropathy because of the lack of adequate rent controls, small number of subjects, and

can be in IgA concur-

versus renal
survival,

compared respectively,

survival

was

similar

0.03). However, actuarial in both groups (overall renal

50%) even though. In the

approximately

Thrombotic Hemolytic
TTP and manifestations
ized by

Thrombocytopenic Uremic Syndrome

HUS probably of the same

Purpura

and
of

subgroup of patients presenting with severe (dialysisdependent) disease, plasma exchange was shown to improve renal survival ( 1 4 of 1 9 treated with plasma exchange versus 4 of 1 1 in the control group were able to discontinue dialysis for 2 months, P = 0.05). The Italian Cooperative Group retrospectively studied 29 patients with HUS/TTP, followed-up for an average of 36.5 months (89). Complete remission was achieved
in 19 of 22 patients who drug (86%) who had been treated with plasma

represent a spectrum disease and are characterwith consump-

plasma
patients infusion

exchange,
(57%) and/or

as

compared
had been therapy.

with
treated

four
with

of seven

thrombotic

microangbopathy

tive thrombocytopenia, anemia, and renal

HUS variant tend

failure

microangbopathic (80,8 1). Patients

and have

hemolytic with the

to be younger

inent
mon

renal

involvement,
and

whereas
frequently

in adulthood

more promis more cornassociated with fever

TTP

and neurologic abnormalities. Postulated pathogenetic mechanisms in HUS/TTP include (1 ) deficiency of anti-thrombotic on fibnlnolytic activity, and (2) the presence of circulating factors that provoke endothelial injury and/or platelet aggregation, and promote microthrombi formation. Bacterial-derived toxins (e.g. Eschenchla coil-derived verotoxin), anti-endothelial antibodies, immune complexes, abnormal von Willebrand multimers, and various toxic agents have been incriminated in the latter regard (see the review by Shepard and Bukowskl, Reference 82). Therapeutic plasma exchange could potentially bene,

One study conducted in children with HUS cornpared plasma exchange with supportive symptomatic therapy but failed to find any significant beneficial effect of plasma exchange (nine of 1 1 [82%] patients treated with plasma exchange and 12 of 22 [55% 1 patients treated with supportive care alone had a creatinine clearance value greater than 80 mL/min per 1.73 m2 at 1 yr of follow-up; P > 0.05) (90). However, cases with creatinine clearance values lower than 60 mL/min/ per 1 .73 m2 and cases with endstage renal failure 1 yr after the acute phase were
found exclusively in In patients the who had not studies been Is

treated
A key

with

plasma

exchange.
aforementioned

question

fit

patients

with

HUS/TTP

by

replacing

deficient

plasma factor and/or removing circulating toxins. In a review of TTP in 1966, Amorosl and Ultmann concluded that the disease Is usually rapidly progressive and almost uniformly fatal (93% fatality rate; 79% within 90 days), and that there was no effective therapy (83). Since the mid- 1960s, numerous treatment
modalities have been attempted, such as steroids,

whether the beneficial effect of plasma exchange is actually the result of the exchange procedure or of plasma Infusion alone. Two randomized controlled trials compared plasma exchange with plasma infusion In patients with HUS/TTP (Table 6) (9 1 ,92). Rock et al. randomized 1 02 adults with TTP to either plasma exchange or plasma infusion with fresh-frozen plasma (9 1 ). In addition, all patients received aspirin and dipyridamole. The outcomes in the two groups were compared 9 days and 6 months after entry Into the
trial. Response to therapy was defined as an Improve-

antiplatelets, heparmn, vincnistine, and splenectomy. With the introduction of plasma Infusion or exchange to the therapeutic regimen, remission rates of greater than 75% have been consistently reported (84-86). These data and the scientific rationale summarized above led to widespread utilization of plasma Infusion or exchange in patients with HUS/TTP. Since the early 1980s, an overall survival rate of 84% has been reported In more than 20 uncontrolled studies that used plasma exchange as the primary therapy (reviewed in Reference 87). Two controlled studies compared plasma exchange with either plasma Infusion or drug therapy alone (steroids, antiplatelet drugs) (Table 6)
(88,89). Both suggested that plasma exchange im-

ment In the platelet count to more than 150 X i09 for 2 consecutIve days. At both 9 days and 6 months, patients who had received plasma exchange had a
higher response rate and a better group, the survival 78%; authors rate than

those
months: sion

who
group,

had
plasma

received
exchange

plasma

infusion

(survival
plasma concluded

at 6
infu-

63%).

Although

proves the remission rate, renal outcome, and mortalIty rate. The French Cooperative Group retrospectively studied 53 patients with HUS/TTP, who had been followed for an average of 38. 1 months (88). Improved

that plasma exchange is preferable to plasma Infusion, interpretation should be guarded, because patients who underwent plasma exchange received three times as much plasma as patients who underwent plasma Infusion. Indeed, Henon did not observe a difference In outcome between plasma infusion and plasma exchange In a smaller multicenter controlled
trial in which 40 adults were randomized to receive

either plasma

daily infusions of 15 or plasma exchange

mL/kg with

of fresh-frozen a mixture of

15

Journal

of the

American

Society

of

Nephrology

377

Plasma

Exchange

and

Kidney

Diseases

>,

Q
C)

a
N. V

0 0 0 C 0 0

C,, 5.) 0 .t
V li

0 .0

:te:
0C

E
0 0
C
SD

a)
C 0 C C)

Lo
oa.0

0)
U)

oo.0

tw

oo.0

oa.0

a
ci) 0 C 0 0

a)

Cl)

LU

E
0
. -

.gi
.C2

0
0 0.

a)V Q ui
.t

a)
::r
U)

E
0
SD

E
Cl)

0
0 C

2aO
t0)t2C

E
E
AC C,,,

ci)
C C 0 0

20aQ
C a)C2.N.C a)00o

0 a)
C

C
Cl)

a)

a)
0 0
Q
-

z
C

0
.

o.0

c2ci.0 C) V

ci)

a)

0 V 0

a)
C C 0 0 1j 0 C

o
a) E0

2
z
C
.

Cl)

2
.U)

2
.-

E
C

ooz
Ciii

N.C

II
o.0

a)
0

ci)

E
0 ci)
C
Cl)

o.0

E a)
a)

ci) 0

a) 0. DV
C -0

C
-;

Ca

:
a) a) 0

a
Dy 0a)
U)

_.-a)
Cl)

E
00

00
<
(l).

E
>

0a)C0

<a (l). <ci

>.

<a

0 #{149}0 C
>-

a) E

SD

0) C C

Cl)

C) x SD
00

.0U)0

h00) a)EC

a)
.Cl)

E z

DC
LU

C,,
. C,,

0
C,)

U)
C)

2
c)
C)

0
C)

0
C)

U)E
20
-C
>,u)

CC 00 06

_.J

e
.

5..
()
U)_

oE
U)

0.0)

-.i5

C C

C)C) C)C

a)Cl) Cl)

0.

Do
a)-

E C
0 a),-

C a)

c)
U).

CU) Oc)

CU)
a)C,1

e:a

a)

ON.

-%

OC)

0 0
C 0) 0

.5
0

_
D

C a) .g V 0 0. .i2 0

.2
U)

o
0.

2U)

0
0.

E
ED
C)0 U)0
SDC 0)
U) C)

U)
>-

0
C,,, >.

C) C) >

C,, 0

0 a) >0 a)0 C a)l;; >0 a)O 0..t C

0 e >D a)0 C

0 V0

0
V0

a
V0.
= V C

a)
>,

:2

L:

o..t
o
a

o..a o
,

E0
0 V

o.2
Co

U)

20

o
0

00

20

0 V

E2

go

20 4-0.
a)U) U)

. .,-

E
U)

Ca)

o::?
%-,

Ce2

0
-

0
a)-,

C,,

5a.(D

<

!: 3:

a)

0 C a)

378

Volume

Number

1996

Madore

et

al

mL/kg

of fresh-frozen

plasma

and

45

mL/kg
largely

of 5%
based

ruricemia,

cryoglobulinemia,

amyloidosis,

light-chain

albumin as replacement In summary, there

fluid (92). is evidence, albeit

deposition, therapeutic

hyperviscosity, agents (96,97).

factor with

infections, Response

and chemoto chernothersurto

on historical controls, that plasma exchange improves renal outcome and mortality in adult patients with HUS/TTP. It is unclear whether this benefit occurs as a result of plasma infusion alone with replacement of a deficient plasma factor or as a result of plasma exchange with removal of a circulating toxic mediator. From a practical viewpoint, It is often necessary to perform plasma exchange to administer the desired amount of plasma as these patients are often oligoaedema. Thus, It seems reasonable to begin plasma infusion promptly upon making the diagnosis and to commence plasma exchange as early as possible thereafter. The following plasma exchange protocol was reported by Rock et at. (9 1): daily plasma exchange is performed for 7 to 14 days, using 4-L exchanges and fresh-frozen plasma as replacement fluid. The results of recent reports suggest that cryosupernatant may be an attractive alternative replacenuric nary and prone to develop hypervolemia and pulmo-

apy Is the major vival. Individuals

that conditions patient tumors unresponsive

chemotherapy

fare

poorly

no matter

what

treatment
protein, and

Is

utilized (98). Serum levels of myeloma type and severity of renal lesions are the

major factors that condition the recovery of renal function (99). Patients with advanced irreparable renal damage (such as severe cast nephropathy with interstitial fibrosis and tubular atrophy) are not likely to respond to any form of therapy. Plasma exchange combined with chemotherapy has been advocated as a means of
rapidly reducing the plasma concentration of my-

eloma proteins and, hence, reducing the filtered load and nephrotoxiclty of these proteins (7). Two randomized controlled trials of plasma exchange in multiple myeloma have been reported (Table 7) ( 100, 1 0 1). Johnson et at. ( 1 00) randomized 21 patients to either chemotherapy plus forced diuresis or to a similar protocol plus plasma exchange and
could detect only a small and nonsignificant benefit

ment
contain

fluid
large

in

some
von

patients,
Willebrand

because
multimers

it does
that

not
have

been implicated in the pathogenesis of TTP (93-95). Response to therapy Is monitored with repeated assessments of platelet counts, lactate dehydrogenase, urine output, and serum creatinine values. Concomitant adjuvant therapy remains highly controversial.

Aspirin, dipyridamole, and splenectomy have

cacy has never been

corticosteroids, been suggested,

demonstrated

but

vincristine, their effi-

despite rapid lowering of plasma myeloma protein. Recovery of renal function (sufficient to discontinue dialysis) was noted in three of seven patients treated with plasma exchange who required hemodlalysis at onset, but in none of the five dialysis-dependent patients in the control group. However, there was no difference in survival between the two groups (25%
survival in both groups at 30 months). In or contrast,

conclusively.

Zucchelll
steroids

et
and

at.

randomized
cyclophosphamide

29

patients
with

to

receive
without

Renal Failure Myeloma

Associated

with

Multiple

plasma atinine
required

exchange values
dialysis.
>

( 1 0 1 ). All patients

Renal failure complicates 3 to 9% of cases of multiplc myeloma and is associated with poor prognosis. Renal impairment is caused by toxicity of myeloma
light chains to renal tubules, although other factors

( 15 and 14 patients, respectively) had severe acute renal failure (cre5 mg/dL [442 tmol/L1) and 24
Thirteen of 15 patients who had

been treated with function (creatinine

LI). In contrast, only

plasma values
two of

exchange recovered renal < 2.5 mg/dL 122 1 mol/

14 patients In the control

can

also

contribute,

including

hypercalcemia,

hype-

TABLE 7. Randomized controlled trials evaluating failure associated with multiple myelomaab Authors
(Ref.)
Patients

the efficacy Number of Plasma Exchanges

3 to 12

of plasma

exchange

in treatment

of acute

renal

on

Concomitant
Therapy (method) Forced diuresis
Steroids (p0) (p0)

Benefit of Plasma Exchange

Renal Outcomec MortaIlty Mortallty:e PE:25%
Confrols:25%

Study

Design

Dialysis at Presentation 12 of 21

Johnson (100)
Zuccheill (101)

et a!.

RandomIzed
controlled trial

21

Dlscont. of dialysis:
PE:43% Controls: 0%

Meiphalan et a!. Randomized

controlled trial

29

24 of 29

Forced dluresis
Steroids (lv+po)

Dlscont. of dIalysis:
PE:84%

MortalIty:t
PE:34%

Cyclo
a

(iv) of dialysis.
and excluded

Controls:18%
patients with chronic

Controls:72%
renal Impairment.

All studies Expressed Expressed Statistically

b
C d

PE. plasma

focused

exchange;

on acute

Cyclo.

or progressive

cyclophosphamide; between

renal

failure

Discont.

associated

of dialysis. discontinuation group

group

with

multiple

myeloma

as percent as percent significant

of patients of patients

on dialysis at presentation. from the entire group.

e
C

No statIstically

significant

dIfference

plasma

exchange
control

benefit

of plasma

exchange

versus

and control (P < 0.05).

group.

Journal

of the

American

Society

of Nephrology

379

Plasma

Exchange

and

Kidney

Diseases

group patients
the

had a similar who required

exchange

outcome. dialysis
group

In the subgroup at presentation, 84%

and 18% in the control

of in

values

plasma

group (P <
months,

recovered enough renal 0.05). After a mean

none of the Improved

function follow-up
patients

to stop period
treated

dialysis of 23
with 1 -yr group

(P < 0.01). Unfortunately, anti-HLA antibody titers returned to baseline levels during the subsequent 4-wk follow-up period. The effect of transient antibody removal on allograft survival was not assessed in this study. In other trials, transplantation was attempted after

plasma
survival

exchange
rate was

had
66%

to resume
in the plasma

dialysis.
exchange

The

versus 28% In the control group (P < 0.0 1 ). These results suggested that plasma exchange improves both renal outcome and patient survival. A similar trend was noted In additional nonrandomized studies and case series by Wahlin et at. Misiani et at. and Pozzi et at. These studies supported the notion that plasma exchange promotes recovery of renal function and lessens the need for chronic dialysis in certain
, ,

prophylactic removal of cytotoxic antibodies by using plasma exchange, with encouraging results ( 105-107). Taube et at. treated 1 7 highly sensitized patients with immunoadsorptbon to remove anti-HLA antibodies before transplantation ( 1 07). Fifteen of the 1 7 patients had positive cross-matches with their

patients

with

myeloma-related

renal

failure

(7,98,99).

Thus, whereas the role of plasma exchange in multiple myeloma remains controversial, there is suffident, albeit limited, data to justify the use of plasma exchange as adjunctive therapy In some patients with myeloma cast nephropathy. However, given the paucity of data, It is Impossible to give fIrm recommenda-

donor when using pre-exchange sera, whereas crossmatches were negative immediately pretransplant after plasma exchange. Only one patient lost his graft because ofrejection (at 1 yr after transplantation). The incidence of allograft dysfunction in the first year after transplant was 3.0 episodes per patient in individuals treated with prednisone and azathioprine, and 1.4
episodes cyclosporine, per patient and in those antithymocyte treated with globulin prednisone, (ATG). De-

spite uate
HLA

these encouraging results, large trials the influence of pretranspiant removal

antibodies by plasma exchange will be

that evalof antirequired

the specifics of therapy. If utilized, it would seem reasonable to perform at least five plasma exchange sessions, using 4-L exchanges and albumin
solution as replacement should be monitored urine output, serum fluid. Response to therapy with repeated assessments of creatinine, and possibly plasma

tions

regarding

before firm cacy of this

Prophylactic

conclusions practice.
plasma

can
exchange

be drawn
has

about
also been

the

effi-

evalu-

ated

as induction

therapy

in highly

sensitized

patients
again in an (103,108). of plasma

myeloma

protein

levels.

in the immediate effort to prevent The results did

postoperative period, early humoral rejection not show a major benefit

Renal

Transplantation

exchange over In a controlled

highly sensitized

conventional antirejection trial, Reisaeter et at.

patients to either

prophylaxis.
randomized standard 17 triple

Over the past 20 yr. plasma exchange has been used in experimental protocols to prevent or treat renal allograft rejection by removing cytotoxic antibodies and other inflammatory mediators. In addition, plasma exchange has been utilized for the prevention and treatment of recurrence of primary renal disease In allografts. Approximately 20% of patients on waiting lists for cadaveric transplantation have high levels of preformed cytotoxic antibodies that render them at high risk for hyperacute and acute allograft rejection (102). In an attempt to solve this problem, plasma exchange and immunoadsorptbon have been assessed before transplantation as a means of removing cytotoxic antibodies. In uncontrolled but compelling studies, plasma exchange significantly reduced the titers of preformed cytotoxic antibodies. However, this benefit
was usually transient and antibody titers rebounded

therapy alone (prednisolone, closporine) or standard therapy

(6 exchanges over a 2-wk period)

cyclophosphamide, plus plasma

cyexchange
was no

( 108).

There

difference in the number of early and in graft survival. In the plasma

seven of nine patients experienced

rejection exchange
rejections

episodes group,
within 4

wk
group,

and
all

five
eight

lost

their
had early who in the

grafts,

whereas
but

in the
no grafts

control
were

rejections

lost.
which plasma

Frasca

et at. reported

similar

results

in a study

in
pe-

13 patients exchange

had undergone prophylactic immediate postoperative

nod as an adjunct to a standard Immunosuppressive protocol (immunosuppressive drugs plus antilymphocyte globulin (ALG) were compared with 1 1 patients

treated with the standard protocol ( 103). SIx rejection episodes were noted in the plasma exchange group compared with eight in the control group. It is discomforting slightly change Four to note that incidence of severe infections higher in patients treated with plasma (eight versus six severe infection episodes). randomized controlled trials have been was exre-

to pretreatment levels over the ensuing weeks (103105). In a representative study, Hakim et at. treated 1 4 transplant candidates with an immunoadsorption system that used Protein A, which selectively removes Ig from plasma ( 104). Three to six procedures were performed on consecutive days. At the end of the treatment course, plasma IgG levels were reduced by
90% 8% and cytotoxic anti-HLA antibodies were

ported on the efficacy of plasma exchange in the treatment of established biopsy-proven acute allograft rejection (Table 8) ( 109-1 12). Blake, Cardella et at. randomized 85 patIents to receive conventional antirejection therapy with or without plasma exchange for

reduced

18-fold

when

compared

with

pretreatment

380

Volume

Number

1996

Madore

et al

TABLE 8. Randomized controlled acute allograft rejectiona

trials evaluating

the efficacy

of plasma

exchange

in treatment
Benefit

of established
of Plasma

Features at Presentation
Authors (Ref.) Study Design

Number
of Plasma Exchanges 5

Concomitant
Therapy (method) Steroids (iv) Graft
Graft

Exchange
Survival
survlval:c 51% NP

N
Diagnosis Graft Vascular
rejection

Biopsy or cellular

Patients
NP

Survival

Blake et (109)

01b

Randomized
controlled trial

85

Acute rejection

PE: 64% Controls:

Allen et a!.
(110)

Randomized controlled fbI

27

Acute

rejectIon

Vascular

rejection

SteroIds Heparin

(iv) (Sc)

Graft

survlval:c 38%

(unresponsive)d

PE: 18% Controls:

Kirubakaran

Randomized controlled trial

24

Acute

rejection

Vascular

rejection

Steroids

(iv)

Graft

survlvai:c 75%

NR

et a!. (1 1 1) Bonomini et a!. (1 12)

(severe)

PE: 33% Controls:

Randomized controlled trial

44

Acute

rejectIon

Vascular

rejection

3 to 7

Steroids Cyclo

(iv) (po)

Graft

survival:e 19%

NR

(unresponsive)

PE: 70% Controls:

0 b C

PE. plasma exchange; Cyclo, cyclophosphamide; Earlier reports of this study included in references

NR. not reported. (1 13.114).

No statistically

significant

difference

between

plasma

exchange

group

and control
(P
<

group.

Acute rejection unresponsive Statistically significant benefit

to conventional therapy with iv steroids. of plasma exchange versus control group

0.05).

treatment of all episodes of acute rejection occurring within the first 3 months of transplantation ( 109, 1 13, 1 14). No statistically significant difference

in

5-yr
groups,

actuarial

graft

survival

was

observed

between
superior plasma trials

graft
exchange
therapy

although there survival in patients

was a trend toward who had undergone

survival than patients who had received pharmacological antirejection therapy alone (graft failure: 7 of 23 [30%I in the plasma exchange group compared with 17 of 21 [81%I in the control group; P < 0.02). It should be noted, however, that these episodes were not typical of acute rejection; the mean interval be-

focused rejection the basis

circulating

exchange P > on the efficacy of with prominent that this process

group, 51%; antiendothelial-cell

(plasma

group,
0.05).

64%;
Three

conventional
smaller

tween do not

transplantation
Thus, the

and
sum

the of

onset therapeutic

of rejection
published

being
to date

1 1 months. change tion. The for

of data or treatment with

plasma exchange in acute vascular inflammation on is mediated in large part by

antibodies (1 10-1 12).

support
the

the
prevention

use

plasma
of acute

exrejecplasma

published

experience

therapeutic

Allen et at. and Kirubakaran 24 patients, respectively,

et at. randomized 27 and to conventional antirejection

exchange controlled

in chronic series and

therapy (iv methyiprednisolone) with or without plasma exchange ( 1 1 0, 1 1 1 ). Renal biopsy was performed in all cases and histological examination confirmed severe vascular rejection. Analysis of shortterm benefit, as evidenced by a reduction in serum creatinine values (6 days and 1 month after entry) and by subsequent graft survival revealed no significant difference between treatment groups. Only one ran-

results reported benefit of plasma progressive deterioration

rejection is limited to a few uncase reports ( 1 03, 1 1 5, 1 1 6). The by Frasca et at. suggest a potential exchange ( 103). Six patients with
of renal function and detect-

able

circulating

cytotoxic

antibodies

were

treated

with

plasma exchange. Improvement noted in one patient (the degree the improvement is not reported

in renal function was and the duration of in the paper, howev-

er).

In three

cases,

stabilization

of the

renal

function

domized
efit of

trial
plasma

suggested

a statistically

significant

ben-

exchange in acute vascular rejection. studied 44 patients with acute rejection and predominantly vascular lesions that had not responded to a 3-day course of iv methylprednisolone ( 1 1 2). Patients were randomized to either a second course of 3 daily doses of iv methylprednlsolone or to plasma exchange. All patients were maintained on

Bonomini

et at.

was noted, and in the remaining two cases, the treatment had no effect on the rate of deterioration in renal function. In general, the results ofother smaller studles have also been disappointing, with improvement in graft function being, at best, modest and usually transient (115,116). A few uncontrolled studies suggest a role for plasma

cytotoxic
during the

drugs
entire

(azathloprine
course of the and a

or exchange

cyclophosphamide)
Those who had

study. higher

been
dence

treated
of graft

with

plasma
failure

had

a lower
actuarial

mcigraft

exchange in the prevention and treatment of recurrent glomerular disease, particularly primary focal segmental glomerulosclerosis, after transplantation (9, 1 17, 1 18). In a representative study by Dantal et at.,

eight

patients

with

recurrent

nephrotic

syndrome

Journal

of

the

American

Society

of

Nephrology

381

Plasma Exchange

and Kidney Diseases

were used
urinary

treated Protein
protein

with A (9).

plasma immunoadsorptlon This technique induced

values in all patients

that a fall in
by an

agulant

or

in fresh-frozen

plasma,

complicates

1 .5 to

excretion

average therapy return Thus, routine

ular

of 82% (P < 0.001), however, the effect of was transient in all but one patient, with a to the preadsorption level within 2 months. to date, there is insufficient data to recommend use of plasma exchange In recurrent glomerdisease.

9% of treatments. to 4% of patients episodes, delayed

hypo-oncotlc fluid

Hypotensive and can be or inadequate

replacement,

episodes triggered volume

or

occur in 0.4 by vasovagal replacement,

anaphylaxis. A

COMPLICATIONS
RIsk:benefit ratio

OF PLASMA
is an

EXCHANGE
consideration

important

recent report suggests that patients who undergo plasma exchange with membrane ifitration can suffer anaphylactoid reactions when on anglotensmn-converting enzyme inhibitors, similar to those described with hemodialysis ( 1 28). Repeated plasma exchange has been proposed to be immunosuppressive and may increase the risk of life-threatening infections in patients pressive by the who are agents. results receiving Although conventional this postulate immunosupIs supported

when discussing the merits of any therapeutic intervention, and the perception that plasma exchange is a benign procedure has undoubtedly contributed to its widespread use for unproven indications. Although plasma exchange is relatively safe when performed by skified clinicians, complications related to either vascular access or the composition of replacement fluids are frequent. Table 9 summarizes the results from several major studies that specifically assessed the type and the incidence of complications related to therapeutic plasma exchange (adapted from Mokrzyckl et at. with permission) ( 1 19-127). The overall incidence of adverse reactions ranges from 1 .6% to 25%, with severe reactions occurring in 0.5% to 3.1%. Hematomas, pneumothorax and catheter infections are the most frequent complications of vascular access, occurring in 0.02 to 4% of treatments. Complicatlons related to the replacement fluids include anaphylactold reactions to fresh-frozen plasma,
coagulopathies induced by inadequate replacement of

of numerous It is noteworthy infectious complications was est randomized controlled

(25, 129-132),

uncontrolled studies that an increase in not observed in the lagtrial of plasmapheresis

of an Inflammatory disease ( 133). As alluded to previously, the potential complications have been highlighted recenfly, with the observation that patients treated with plasma exchange for lupus nephritis tended to have a worse outcome than those treated with conventional therapy. treatment

CONCLUSION
In summary, there is some most part uncontrolled, that useful adjunct to conventional
in the treatment of antl-GBM

evidence, albeit plasma exchange Immunosuppression

nephrltls, particularly

for the is a
if

clotting factors, transmission other infections. Symptomatic from citrate infusion, either

of viral hepatitis, and hypocalcemia resulting as the treatments antico-

instituted early dent. In contrast, controlled trials little additional sive regimens

before patients are dialysis-depenthe results of several randomized suggest that plasma exchange Is of benefit to standard immunosuppresIn the treatment of pauci-immune

TABLE 9. Major (adapted from

series evaluating the incidence of complications Mokrzycki et a!. (1 19) with permission) Number
Treatments

related

to therapeutic Adverse Reactions

plasma

exchange

Authors (Ref.)
Centrifugal System

of
Total Mild0

Moderateb

Severec

Deaths

Borberg
Aufeuvre Ziselman Rossi

(1 20)
et a!. (121)

205 3086 1389

13%
22% 1 .6%

0
4.8% 0.4%

11.2%
16.4% 0.6%

2%
0.6% 0.5%

0
0.1% 0.1%

et a!. (1 22)

Fabre et a!. (123)

et a!. (1 24)

578
926 306 120 (1 19) 699

25%
17.3%

23%d
7.6%

23%d
6.5%

1.5%
3.1%

Membrane-Based System Sprenger et a!. (125) Samtleben et a!. (1 26)

Mokrzycki and Kaplan

4.2%
17.5%

0
5%

4.2%
12.5%

0
0

0
0

9.7% 12%
and without
significance.

5.4% 9%
consequence.

2.4% 3%

0.7% 0.5%

0 0

Both Types of System Sutton et a!. (127)

a
b C

5235
defined
as symptoms

Mild

Moderate

adverse

adverse

reactions

reactions

defined

as easily treatable

of little

clinical

prolonged

Severe adverse reactions defined as potentially life-threatening. Mild and moderate reactions were grouped together In this study.

382

Volume

Number

1996

Madore

et al

RPGN, Although sively

except
In mixed

in severe
essential

dialysis-dependent
plasma exchange cryoglobulinemla, Is used

patients.
extenthe case

12.

therapeutic

Low A, Hotze A, Krapf F, et aL: The ance function of the reticuloendothelial tients with immune complex mediated and after therapeutic plasmapheresis.
1985:5:69-72.

nonspecific system diseases Rheumatol

clearin pabefore

mt

for

its

use

in

this

disorder

is largely
for and

unproven

and
13.

warrants trials do

further investigation. not support a role

Data from therapeutic suggest that

controlled plasma plasma

Bystryn JC, GraS MW, Uhr JW: Regulation of antibody formation by serum antibody. Removal of specific antibody by means of exchange transfusion. J Exp Med
1970:132:1279-1287.

exchange
GN tients. such The

in other
as lupus

forms
nephrltis,

ofimmune affect
of plasma

complex-mediated
14.

exchange to have

may

adversely

outcome
exchange

in some
appears

paand
to be

introduction in patients with

dramatically definitively
of a circulating

improved
HUS/TTP.

renal this
factor

outcome
It remains

mortality

determined
the removal

whether
toxic

benefit

reflects
of

or infusion

normal change

plasma components. The in myeloma cast nephropathy

role

of plasma exremains controjustified, given argue against of preventing

versial, but further clinical trials are the actual data. Most controlled studies the use of plasma exchange as a means

or treating allograft rejection, although its potential role In preparing highly sensitized patients for transplantation has not been adequately evaluated. The more widespread application of prospective randomized controlled clinical trials should help to better define the true value of plasma exchange in our
armamentarium of treatments for renal diseases.

ACKNOWLEDGMENTS
F. Madore Is a recipient of DK a research fellowship from the Medical

Research
Institutes
Advisory

Council
of Health

of Canada.
Grant

HR.
44380

Brady

is the

recipient

of National

and an award

from the Research

Euler HH, Schroeder JO, Zeuner RA, Teske E: A randomized trial of plasmapheresis and subsequent pulse cyclophosphamide In severe lupus: Design of the LPSG trial. Int J Artif Organs 1991 ; 14:639-646. 15. Haki.m RH, Siami A. Plasmapheresis. In: Daugirdas JT, Ing T, Eds. Handbook of Dialysis. 2nd Ed. Boston: Little, Brown: 1994:219-241. 16. Samtleben W, Gurland H. Plasma exchange: Principles and practice. In: Maher JF, Ed. Replacement of Renal Function by Dialysis. 3rd ed. Dordrecht: Kluwer Academic Publishers; 1989:460-474. 1 7. Hakim RM, Schulman G, Churchill WJ, Lazarus JM: Successful management of thrombocytopenia. microangiopathic anemia, and acute renal failure by plasmapheresis. Am J Kidney DIs 1985:5:170-176. 18. Tisher CC, Brenner BM. Renal pathology with clinical and functional correlations. Philadelphia: J.B. Lippincott: 1989. 19. Wilson CD, Dixon FJ: Anti-glomerular basement membrane antibody-Induced glomerulonephritis. Kidney Int 1973:3:74-89. 20. Rees AJ. Goodpastures syndrome. In: Glassock RJ, Ed. Current Therapy in Nephrology and Hypertension. St. Louis: Mosby-Year Book; 1992:173-178. 2 1 . Johnson JP, Moore JJ, Austin HB, Balow JE, Antonovych TT, Wilson CB: Therapy of anti-glomerular basement membrane antibody disease: Analysis of prognostic significance of clinical, pathologic and treatment factors. Medicine (Baltimore) 1985:64:219-227. 22. Simpson U, Doak PB, Williams LC, et at.: Plasma exchange in Goodpastures syndrome. Am J Nephrol
1982:2:301-311.

Group

of the

Veterans

Administration.

23.

REFERENCES
1 . Shumak N Engl
2.

Couser WG: Classification,

Am J Kidney

Rapidly progressive glomerulonephritis: pathogenetic mechanisms, and therapy.

Dis 1988; 11:449-464.

KH, Rock GA: Therapeutic J Med 1984;310:762-71.

plasma

exchange.

24.

3.

4.

5.

6.

Pineda AV: Applications of therapeutic apheresis. Mayo Clin Proc 1994;69:893-894. American Medical Association: Panel on Therapeutic Plasmapheresis: Current status of therapeutic plasmapheresis and related techniques. JAMA 1985;253:819825. Klein H, Balow J, Dau P, et aL: Clinical applications of therapeutic apheresis. J Clin Apheresis 1986:3:1-92. Leitman S. Kucera E, McLeod B, et al. Guidelines for Therapeutic Hemapheresis. Bethesda, MD: American Association of Blood Banks: 1992. Strauss RG, Ciavarella D, Gilcher RO, et at.: An overview of current management. J Clin Apheresis 1993;8:
195-210.

25.

Lockwood CM, Boulton JJ, Lowenthal RM, Simpson U, Peters DK: Recovery from Goodpastures syndrome after Immunosuppressive treatment and plasmapheresis. Br Med J 1975;2:252-254. Lockwood CM, Rees AJ, Pearson TA, Evans DJ, Peters DK, Wilson CB: Immunosuppression and plasmaexchange in the treatment of Goodpastures syndrome.
Lancet 1976:1:71 1-7 15.

26. 27.

28. A, Lofvenberg E, Hoim J: Improved myeloma with renal failure. Acta

1:205-209.
survival in

7.

Wahlin multiple
1987;22

Med

Scand 29. 30.

8. Kelton JG, Neame PB, Walker I, et al.: Thrombotic thrombocytopenic purpura: Mechanism for effectiveness of plasmapheresis lAbstracti. Clin Res 1979:27:
299A.

9. Dantal J, Bigot E, Bogers W, et at.: Effect of plasma protein adsorption on protein excretion in kidneytransplant recipients with recurrent nephrotic syndrome. N Engl J Med 1994:330:7-14. 10. Byrnes JJ, Khurana M: Treatment of thrombotic thrombocytopenic purpura with plasma. N Engl J Med
1977;297: 1 1 . Lockwood 1386-1389.

Pusey CD, Lockwood CM: Plasma exchange and immunosuppressive drugs In the management of severe gbmerubonephrltis. Prog Clin Blob Res 1982:106:91-104. Peters DK, Rees AJ, Lockwood CM, Pusey CD: Treatment and prognosis in antibasement membrane antibody-mediated nephritis. Transplant Proc 1982:14: 5 13-52 1. Pusey CD, Lockwood CM, Peters DK: Plasma exchange and immunosuppressive drugs In the treatment of gbmerubonephritls due to antibodies to the gbomerular basement membrane. Int J Artif Organs 1983:1:15-18. Pusey CD, Lockwood CM: Antibody mediated glomerubonephritis: Clinical and experimental studies. J Clin Apheresis 1985:2:396-404. Ewert BH, Jennette JC, Falk RJ: Anti-myeboperoxldase antibodies stimulate neutrophils to damage human
endothelial cebls. Kidney Int l992;41:375-383.

CM,

Woriledge

S.

Nicholas

A,

Cotton

C,

Peters
patients exchange.

DK:

Reversal

of impaired

splenic

function
by plasma

in

with nephrltis or vasculitis (or both) N Engl J Med 1979;300:524-530.

3 1 . Falk JU, Terrell R, Charles LA, et at.: Anti-neutrophil cytoplasmic autoantibodies Induce neutrophibs to degranulate and produce oxygen radicals In vitro. Proc NatI Acad Sd USA 1990:87:4114-4119. 32. Kain R, Matsui K, Exner M, et at.: A novel class of autoantigens of anti-neutrophil cytoplasmic antibodies in necrotizing and crescentic glomerubonephritis: The bysosomal membrane glycoprotein h-bamp-2 in neutrophil granubocytes and a related membrane protein in

Journal

of the

American

Society

of Nephrology

383

Plasma

Exchange

and

Kidney

Diseases

33.

34.

35.

36.

37.

endothellal cells. J Exp Med 1995:181:585597. Gl#{246}ckner WM, Sieberth HG, Wichmann HE, et al.: Plasma exchange and Immunosuppresslon In rapidly progressive glomerulonephrltls: A controlled, multicenter study. ClIn Nephrob 1988;29: 1-8. Cole E, Cattran D, Magil A, et aL: A prospective randomized trial of plasma exchange as additive therapy In idiopathic crescentic glomerubonephrltls. The Canadian Apheresis Study Group. Am J Kidney Dis 1992:20:26 1269. Pusey CD, Rees AJ, Evans DJ, Peters DK, Lockwood CM: Plasma exchange In focal necrotizing glomerulonephrltis without anti-GBM antibodies. Kidney mt 1991; 40:757-763. Rifle G, Dechelette E: Treatment of rapidly progressive glomerulonephrltls by plasma exchange and methylprednisolone pulses. A prospective randomized trial of cycbophosphamlde. Interim analysis. The French Cooperative Group. Prog Clin Blob Res 1990;337:263-267. Mauri JM, Gonzales MT, Poveda R, et at.: Therapeutic plasma exchange In the treatment of rapidly progressive gbomerulonephrltis. Plasma Ther & Transfus Techglomerubar nob 1985;6:587-591.

52. 53.

54.

55.

56.

Euler Iffi, Schroeder JO: Plasmapheresis for lupus nephritis. Lupus Plasmapheresis Study Group. N Engl J Med 1992;327:1029-1030. Schroeder JO, Euler HH, L#{246}ffler H: Synchronization of plasmapheresis and pulse cyclophosphamide In systemic lupus erythematosus. Ann Int Med 1987; 107: 344-346. Barr WG, Hubbel EA, Robinson JA: Plasmapheresis and pulse cyclophosphamide In systemic lupus erythematosus. Ann mt Med 1988:108:152-153. Huston DP, Bressler RB, Suki WN, et at.: A controlled study of pulse cyclophosphamide and plasmapheresis therapy of lupus nephrltis [Abstractl. Arthritis Rheum 1988;3llSuppl 55]:31A. Ziellnskl C, Mueller C, Smolen J: Plasmapheresis in the treatment of systemic lupus erythematosus: A controlled study. Acta Med Austriaca 1988; 15:155-158.
Brouet JC, Clauvel JP, Danon F, Klein M, Seligmann

57.

58.

38. 39.

Clough JD, Calabrese LH: Theoretical mune complex removal by plasmapheresis. Ther Transfus Technol 1981:2:73-81. Derksen RH, Schuurman HJ, Gmelig Verroust P, Kater L: Circulating Immune systemic lupus erythmatosus: 1elatlon tivity and the effect of plasma exchange.
Transfus Technol 1985:6:327-335.

aspects

of ImPlasma

59.

Meyling FHJ, complexes in to disease acPlasma Ther &

60.

M: Biologic and clinical significance of cryoglobulins. A report of 86 cases. Am J Med 1974;57:775-788. DAmico G, Fornasieri A: Cryoglobullnemlc glomerulonephrltls: A membranoproilferative glomerulonephritls induced by hepatitis C virus. Am J Kidney Dis 1994:25: 361-369. Frankel AR, Singer DR, Winearis CG, Evans DJ, Rees AJ, Pusey CD: Type II essential mixed cryoglobulinaemia: Presentation, treatment and outcome in 13 patients. Q J Med l992;82:101-124. Gorevic PD, Kassab HJ, Levo Y, et at.: Mixed cryoglobulinemla: Clinical aspects and long-term follow-up of 40 patIents. Am J Med 1980; 69:287-308.

Jones LW Robinson MF, Parciany RK, Layfer LF, McLeod B: Therapeutic plasmapheresis In systemic lupus erythematosus. Effects on immune complexes and antibodies to DNA. Arthritis Rheum 1981:24:11131120. 4 1 . Jones JV, Cumming RH, Bucknall RC, Asplin CM: Plasmapheresls in the management of acute lupus erythematosus. Lancet 1976; 1 :709-7 1 1. 42. Jones JV, Cumming RH, Bacon PA, et at.: Evidence for a therapeutic effect of plasmapheresis In patients with systemic lupus erythematosus. Q J Med 1979:48:55540.
576.

43.

44. 45. 46.

Leaker BR, Becker GJ, Dowling JP, Kincaid-Smith P: Rapid Improvement in severe lupus glomerular lesions following Intensive plasma exchange associated with immunosuppresslon. Clin Nephrol 1986;25:236-244. Morlconi L, Fern C, Fanara G, et al.: Plasma exchange in the treatment of bupus nephrltis. mt J Artif Organs
1983; 1:35-38.

VangellstaA,

Frasca

G, Orsi

C, Sermasi

Bonomini
nephritls.

V: Short-term

plasmapheresis

G, Zucchelli P, in acute bupus

6 1 . Geltner D, Kohn RW, Gorevic P. Franklin EC: The effect of combination therapy (steroids, immunosuppressives, and plasmapheresis) on 5 mixed cryoglobulmnemia patients with renal, neurologic, and vascular Involvement. Arthritis Rheum 1981:24:1121-1127. 62. Berkman EM, Orlin JB: Use of plasmapheresis and partial plasma exchange in the management of patients with cryoglobulinemla. Transfusion l980;20: 171-178. 63. Fern C, Moriconi L, Gremignal G, et at.: Treatment of the renal Involvement In mixed cryoglobuilnemia with prolonged plasma exchange. Nephron 1986:43:246253. 64. Singer DR, yenning MC, Lockwood CM, Pusey CD: Cryoglobulinaemia: Clinical features and response to treatment. Ann Med Interne 1986:137:251-253. 65. Sinlco RA, Fornasieri A, Florini G, et at.: Plasma exchange in the treatment of essential mixed cryoglobulinemla nephropathy. Long-term follow up. mt J Artif Organs l985;2:15-18. 66. Valbonesi M, Montani F, Mosconi L, Florio G, Vecchi C: Plasmapheresls and cytotoxic drugs for mixed cryoglobulinemla. Haematologla (Budap) 1984; 17:341-351.

47.

48.

49.

mt J Artif Organs 1983; 1:43-46. LeWIS E1, Hunsicker LG, Lan SP, Rohde 1W, Lachin JM: A controlled trial of plasmapheresis therapy in severe lupus nephritis. The Lupus Nephrltis Collaborative Study Group. N Engi J Med 1992;326: 1373-1379. Wel N, KlIppel JH, Huston DP, et at.: Randomized trial of plasma exchange In mild systemic lupus erythematosus. Lancet 1983;l:17-22. Derksen RH, Hene RJ, Kallenberg CG, Valentijn RM, Kater L: Prospective multicentre trial on the short-term effects of plasma exchange versus cytotoxic drugs in steroid-resistant lupus nephritis. Neth J Med 1988;33: 168-177. The French Cooperative Group. A randomized trial of

67.

LAbbate effects essential

1985;2:

A, Maggiore Q Cacca.mo A, et aL: Long term of cryoapheresis and cytostatic treatment in mixed cryoglobulinemia. mt j Artif Organs
19-22.

68. 69. 70.

Schena exchange
Lab

FP,
in

Manno C, the treatment

Dimonte D, et ofcryoglobullnemia. Kidney and Int treatment

A review.

at.:

PlasmaRic Clin

1983:13:133-140.

Galla JH: IgA nephropathy. 387. Glassock RJ: Natural history

proliferative glomerulonephrltls:

1995;47:377of primary
Kidney Int L: IgA of high 1980;

Suppl

l985;lSuppl

171:5136-5142.

plasma Plasma
50.

exchange Ther

in severe
Methodology

acute
and

systemic
interim

lupus
analysis.

cry-

7 1 . Lopez-Trascasa M, Egido J, Sancho J, Hernando glomerulonephrltls (Bergers disease): Evidence serum levels of polymeric IgA. Clin Exp Immunol
42:247-254.

thematosous:

& Transfus

Technol

1985;6:535-539.

72.

5 1.

Clark WF, Cattran DC, Balfe JW, Williams W, Lindsay RM, Linton AL: Chronic plasma exchange in systemic lupus erythematosus nephrltls. Proc Eur Dial Transplant Assoc l983;20:629-635. Bystryn JC, Schenkheln I, Uhr JW. A model for regubatlon of antibody synthesis by serum antibody. In: Amos B, Ed. Progress in Immunology. New York: Academic Press; 1971;627-636.

Coppo R, Basolo B, Roccatello D, Piccoli G: Plasma exchange in primary IgA nephropathy and HenochSchOnleln. Plasma Ther & Transfus Technol 1985:6:
705-723.

73. 74.

Chalopin JM, IgA nephropathles

13:45A.

Tanter

Y, Rifle lAbstract].

G: Plasma exchange Eur J Clin Invest D, et at.: Plasma IgA nephropathy.

and 1983: Int cxJ

Coppo change

R, Basolo B, Roccatello In progressive primary

384

Volume

Number

1996

Madore

et at

75.

Artif Organs 1985;2:55-58. Coppo R, Basolo B, Giachino 0, et at.: In a patient with rapidly progressive nephropathy: removal oflgA-containing mune complexes and clinical recovery. 40:488-490.
Hene RJ, Kater L: Plasmapheresis in

95. Plasmapheresis Idiopathic IgA circulating tinNephron 1985; 96.

nephrltis associ-

Obrador Hanrahan therapy

GT, Zeigler ZR, Shadduck JB: Effectiveness of In refractory and chronic

purpura. Am

RK, Rosenfeld CS, the cryosupernatant relapsing thrombotic

J Hematol 1993:42:

thrombocytopenic

76.

ated IgA 77.

with Henoch-Schonlein nephropathy. Plasma

purpura and Ther & Transfus

In primary Technol

97. 98. 99.

1983:4:165-173.

Kauffmann progressive and the

Clin

RH, Houwert DA: Plasmapheresis in rapidly Henoch-Schoenlein gbomerulonephrltls effect on circulating IgA immune complexes.
1981;l6:l55-160.

Nephrol

78.

Lai KN, Lai FM, Leung

79. 80.

AC, Ho CP, Vallance OJ: Plasma exchange in patients with rapidly progressive idiopathic IgA nepliropathy: A report of two cases and review of literature. Am J Kidney Dis 1987; 10:66-70. Nicholls K, Becker G, Walker R, Wright C, KincaidSmith P: Plasma exchange in progressive IgA nephropathy. J Chin Apheresis 1990:5:128-132. Byrnes JJ, Moake JL: Thrombotic thrombocytopenlc purpura and the haemolytlc-uraemia syndrome: evolvIng concept ofpathogenesls and therapy. Clin Haematol
1986; 15:413-442.

2 17-220. DeFronzo RA, Cooke CR, Wright JR. Humphrey RL: Renal function In patients with multiple myeboma. Mcdicine (Baltimore) 1978:57:151-166. Fang LST: Light chain nephropathy. KIdney Int 1985: 27:582-592. Misiani R, Tlraboschi G, Mingardi G, Mecca G: Management of myeboma kidney: An anti-light-chain approach. Am J Kidney Dis 1987; 10:28-33. Pozzi C, Pasquall 5, Doninl U, et at.: Prognostic factors and effectiveness of treatment in acute renal failure due to multiple myeloma: A review of 50 cases. Report of the Italien Renal Immunopathobogy Group. Clin Nephrol
1987;28: 1-9.

100.

Johnson WJ, Kyle PA, Pineda AA, OBrien PC, Holley KE: Treatment of renal failure associated with multiple myeloma. Plasmapheresis, hemodialysis. and chemotherapy. Arch Intern Med 1990; 150:863-869.

10 1 .

Zucchelli

P, Pasquali
in

5, Cagnoli
acute renal

L, Rovinetti
failure due

C: Plasma
to light

8 1 . Remuzzi G: HUS and TTP: Variable expression of a single entity. Kidney Int l987;32:292-308. 82. Shepard Ky, Bukowski RM: The treatment of thrombotic thrombocytopenic purpura with exchange transfusions, plasma infusions, and plasma exchange. Semm Hematol 1987;24:l78-193. 83. Amorosi EL, Ultmann JE: Thrombotic thrombocytopenic purpura: Report of 16 cases and review of the literature. Medicine (Baltimore) l966;45: 139-159. 84. Bukowski RM, Hewlett JS, Refiner RR, Groppe CW, Weick JK, Livingston RB: Therapy of thrombotic thrombocytopenic purpura: An overview. Semin Thromb Hemostasis 1 98 1 7:1-8. 85. Ridolfi RI, Bell WR: Thrombotic thrombocytopenic purpura: Report of 25 cases and review of the literature. Medicine (Baltimore) 198 1 ;60:4 13-428. 86. Pisciotto P, Rosen D, Silver H, et aL: Treatment of
thrombotic thrombocytopenic

102, 103. 104.

exchange therapy chain myeloma. 1984:30:36-39. Keown PA: The

Impact 74-78. and

Trans highly

Am

Soc

Artif

Intern

Organs Etiology,
1987:19:

sensitized
Transplant

patient:
Proc

management.

Frasca GM, Martella D, Vangelista A, Bonomini V: Ten years experience with plasma exchange In renal transplantation. Int J ArtifOrgans 1991:14:51-55. Hakizn RM, Milford E, Himmelfarb J, Wingard R, Lazarus JM, Watt RM: Extracorporeal removal of anti-HLA antibodies in transplant candidates. Am J Kidney Dis
1990; 16:423-431.

105.

Palmer

A, Taube

D, Welsh

K, Bewick

M, Gjorstrup

P,

Thick
poreal 106. 107.

M: Removal
immunoadsorptlon

of anti-HLA

antibodies
renal

by extracortransplanta-

to enable

tion. Lancet

l989;l:10-12.

plasma 87.

exchange

and

review

purpura. Evaluation of the literature. Vox

of Sang

1983:45:185-196.

88.

89.

Onundarson PT, Rowe JM, Heal JM, Francis CW: Response to plasma exchange and splenectomy in thrombotic thrombocytopenic purpura. A 10-year experience at a single institution. Arch Intern Med 1992;152:79l796. French Cooperative Study Group for Adult HUS: Adult hemolytic uremic syndrome with renal microangiopathy. Outcome according to therapeutic protocol in 53 cases. Ann Med Interne 1992; l43lSuppl 11:27-32. lacone A, Dragani A, Di 0G. et at.: Thrombotic thrombocytopenlc purpura tive retrospective 1986:71:39-43. (TTP) study on treatment: 29 cases. Italian cooperaHaematologica

108.

Taube DH, Williams DG, Cameron JS, et aL: Renal transplantation after removal and prevention of resynthesis of HLA antibodies. Lancet 1984; 1:824-828. Taube D, Palmer A, Welsh K, Bewick M, Snowden 5, Thick M: Removal of antl-HLA antibodies prior to transplantation: An effective and successful strategy for highly sensitized renal albograft recipients. Transplant Proc l989;694-695. Relsaeter AV, Fauchald P. Leivestad T, et at.: Plasma exchange in highly sensitized patients as induction therapy after renal transplantation. Transplant Proc
1994:26:1758.

109.

Blake P, Sutton D, Cardella CJ: acute renal transplant rejection. 1990:337:249-252.

Allen Slapak NH, Dyer P, M: Plasma Geoghegan exchange T, in

Plasma Prog

exchange Clin Biol

in Res

1 10.

Harris K, acute renal

Lee HA, albograft

rejection.
425-428.

A controlled

trial.

Transplantation

1983:35:

90.

9 1.

Gianviti A, Perna A, Caringella A, et at.: Plasma cxchange in children with hemolytic-uremic syndrome at risk of poor outcome. Am J Kidney Dis 1993:22:264266. Rock GA, Shumak KH, Buskard NA, Ct at.: Comparison of plasma exchange with plasma infusion In the treatment of thrombotic thrombocytopenic purpura. Canadian Apheresis Study Group. N Engl J Med 1991:325:
393-397.

1 1 1 . Kirubakaran MG, Disney Mathew TH: A controlled

treatment of renal allograft

AP, Norman J, Pugsley DJ, trial of plasmapheresis In the

rejection. Transplantation

198l;32:
1 12. Bonomini

164-165.

92. 93.

94.

Henon P: Treatment of thrombotic thrombopenlc purpura. Results of a multicenter randomized clinical study. Presse Med 1991:20:1761-1767. Byrnes JJ, Moake JL, Klug P. Penman P: Effectiveness of the cryosupernatant fraction of plasma in the treatment of refractory thrombotic thrombocytopenic purpura. Am J Hematol l990;34: 169-174. Dayoan EJ, Dixon AC, Rizzo E, Nakainura JM: Refractory thrombotic thrombocytopenic purpura: Successful
treatment by plasmapheresis with plasma cryosuper-

1 13.

V, Vangelista A, Frasca GM, Di FA, Liviano DG: Effects of plasmapheresis In renal transplant rejection. A controlled study. Trans Am Soc Artlf Intern Organs 1985:31:698-703. Cardella CJ, Sutton DM, Katz A, et at. In: Zurukzoglu w, Papadimltrlou M, Pyrpasopoulos M, Eds. Eighth International Congress of Nephrobogy. Athens. Greece: Karger: 1981:681.

1 14.

Cardella
trial

CJ,

Sutton

DM,

Katz
plasma

A, et

evaluating

intensive

at.: exchange

A controlled in renal

1 15.

transplant recipients. Proc Eur Dial Transplant Assoc 1980; 17:429-434. Hendrick A, Wallin JD, ONeill WJ: Plasma exchange in chronic renal allograft rejection. Transplantation 1984:
37:318-319.

natant.

Haematologlca

1993:78:389-392.

Journal

of the

American

Society

of Nephrology

385

Plasma

Exchange

and

Kidney

Diseases

1 16. 1 17.

Rifle

G, Chalopin

JM,

Turc

JM,
allograft

et

In the treatment of renal plant Proc 1979:11:20-26.

at.: Plasmapheresis rejections. Trans-

separation:

Complications

and

monitoring.

Artif D, Gurland

OrH.
with

126.

gans 1984:8:360-363. Samtleben W, Hillebrand

Membrane plasma separation:

G, Krumme
Clinical

Li PK, Lal FM, Leung CB, Lui SF, Wang A, La! KN: Plasma exchange in the treatment of early recurrent focal gbomerulosclerosis after renal transplantation. Report and review. Am J Nephrol 1993; 13:289-292.
Mowry J, Marik J, Cohen A, Hogg R, Sahney S, Et-

experience

1 18.

tenger
1 19.

R: Treatment of recurrent merulosclerosls with high-dose plasmapheresis. Transplant Proc

Mokrzyckl MH, Kaplan AA:

focal segmental cyclosporine A 1993:25:1345-1346.

plasma

gband
cx-

127.

more than 120 plasma exchanges. In: Seiberth H, Ed. Plasma Exchange. Stuttgart, Germany: FK Schattauer Verlag; 1980:23-27. Sutton D, Nair R, Rock G: The Canadian Apheresis
Study Group: Complications Transfusion 1989;29: 124-127. of plasma exchange.

Therapeutic

128.

Owen
with

HG,
use of

Brecher

ME:

Atypical

reactions
enzyme

associated
inhibitors

change:
Dis

Complications

and

management.

Am J Kidney 129.

anglotensin-converting

1994:23:817-827.

1 20.

Borberg Gurland Plasma

1980:201. Aufeuvre

H. Problems of plasma HJ, Heinze V. Lee Exchange. New York:

J, Morin-Hertel

exchange therapy. In: HA, Eds. Therapeutic Springer-Verlag: 191;

M, Lefioch A,

and apheresis. Transfusion Hind CR, Paraskevakou

1994:34:891-894. H, Lockwood CM, Evans

DJ,

Peters
130.

DK,

Rees

AJ:

Infectious
5,

complications
gbomeruloneA, Cotton Nicholas

with
C,
cresP, syn-

12 1.

F, Cohen-Solal

Baudelot
122. 123.

J. Hazards

of plasma

exchange.

In: Sieberth

plasmapheresis in rapidly phrltls. JAMA 1980:244:2423-2426. Lockwood CM, Woriledge Peters DK: Plasma-exchange
in the treatment of fulminating

progressive

and

124.

H, Ed. Plasma Exchange. Stuttgart, Germany: FK Schattauer Verlag; 1980:149-157. Ziselman E, Bongiovanni M, Wurzel H: The complications of therapeutic plasma exchange. Vox Sang 1984: 46:270-276. Fabre M, Andreu G, Mannoni P. Some biological modificatlons and clinical hazards observed during plasma exchanges. In: Selberth H, Ed. Plasma Exchange. Stuttgart, Germany: FK Schattauer Verlag; 1980:143-148. Rossi P, Cecchlni L, Minichella G, et at.: Comparison of the side effects of therapeutic cytapheresis and those of other types of hemapheresis. Haematologia 199 1 ;76:
75-80.

Immunosuppression immune-complex GJ, Kincaid-Smith Goodpastures

centlc 1 3 1 . Walker
Craswell

nephritls.

Lancet

1977:1:63-67.
in

RG, dApice AJ, Becker PW: Plasmapheresis

132.

drome with renal failure. Med J Aust 1977:1:875-879. Wing EJ, Bruns FJ, Fraley DS, Segel DP, Adler 5: Infectious complications with plasmapheresis in rapIdly progressive gbomerubonephritis. JAMA 1980:244:
2423-2426.

133.

Pohi MA, Lan SP, Ben T: Plasmapheresis does increase the risk for infection in Immunosuppressed patients with severe lupus nephritis. The Lupus phritls Collaborative Study Group. Ann Intern
1991;! 14:924-929.

not NeMed

125.

Sprenger

K, Rasche

H, Franz

H: Membrane

plasma

386

Volume

Number

1996