Kidney International, Vol. 58, Suppl. 76 (2000), pp.
S-148–S-155
Adsorption in sepsis
CLAUDIO RONCO, ALESSANDRA BRENDOLAN, MAURIZIO DAN, PASQUALE PICCINNI,
RINALDO BELLOMO, CONCETTA DE NITTI, PAOLA INGUAGGIATO, and CIRO TETTA
Renal Research Institute, New York, New York, USA; Department of Nephrology and Intensive Care, St. Bortolo Hospital,
Vicenza, Italy; Intensive Care Unit, Austin and Repatriation Medical Center, Melbourne, Australia; “Baldi and Riberi” Central
Laboratories, Molinette Hospital, Turin, and Clinical and Laboratory Research Department, Bellco S.P.A., Mirandola, Italy
Adsorption in sepsis. The pathophysiology of sepsis offers a
highly complicated scenario. In sepsis, endotoxin or other
gram-positive–derived products induce a complex and dynamic
cellular response, giving rise to several mediators known to be
relevant in the pathogenesis of septic shock such as specific
mediators, substances responsible for up- or down-regulation
of cytokine receptors and cytokine antagonists, inactivators of
translational or transductional pathways, and precursor mole-
cules. In this review, we delve into some new concepts stem-
ming up from the use of sorbents in continuous plasma filtra-
tion. Nonspecific simultaneous removal of several mediators
of the inflammatory cascade have led to improved outcomes
in animal models of septic shock and to improved hemodynam-
ics in a pilot clinical study. It seems of great importance to
explore all possible treatment techniques that may have a direct
impact on circulating mediators of sepsis and that also may
interfere with the imbalance between proinflammatory and
anti-inflammatory substances in the critically ill patient with
multiple organ failure. In this view, the application of sorbents
appears to open new and interesting therapeutic options. The
search for innovative treatments specifically targeted to the
special needs of the critically ill patients seems therefore more
important than the attempt to adjust concepts and technologies
that are normally applied to patients with chronic renal failure.
Sepsis is the leading cause of acute renal failure (ARF)
and mortality in intensive care units [1–3]. It generally
develops as a result of the host response to infection [4].
The pathogenesis of sepsis represents a complex mosaic
of interconnected events in respect to which therapeutic
strategies remain elusive. Sepsis may be considered a
form of severe systemic inflammation caused by local
and systemic effects of circulating proinflammatory me-
diators [5]. Although several therapeutic attempts have
targeted specific components in the proinflammatory
septic cascade, no improvement in survival has been
obtained in large-scale clinical trials focusing on specific
molecules [6]. Components of gram-negative bacteria
cell wall, such as the lipid A-containing lipopolysaccha-
Key words: sorbent, cytokines, renal replacement therapy, plasma fil-
tration, inflammatory cascade.
 2000 by the International Society of Nephrology
S-148
ride (LPS), trigger a global response that involves both
cellular and humoral pathways with the generation of pro-
inflammatory and anti-inflammatory mediators [5, 7, 8].
These include a major group of middle molecular size
(5 to 30 kD) peptides, collectively called cytokines, phos-
pholipase A2-dependent products such as platelet-acti-
vating factor (PAF), leukotrienes, and thromboxanes,
and complement-derived activated products such as C3a,
C5a, and their desarginated products. Other agents
play a role in the pathophysiology of sepsis such as sur-
face-expressed and soluble adhesion molecules, kinins,
thrombin, myocardial depressant substance(s), ␤-endor-
phin, and heat shock proteins. In normal conditions, the
biological activity of these mediators is under the control
of specific inhibitors that may act at different levels.
In sepsis or systemic inflammatory response syndrome
(SIRS), the homeostatic balance is altered, and a pro-
found disturbance of relative production of different me-
diators may be observed. The pathogenesis of sepsis
might therefore be described by a sudden and severe
imbalance of proinflammatory and anti-inflammatory
factors in the host. The concept of sepsis as a simply
proinflammatory event has been consequently chal-
lenged. Recent studies on knockout mice have shown
that deficiency of the gene encoding for intercellular
adhesion molecule-1 renders mice resistant to the lethal
outcome of high-dose endotoxin shock [9]. In these con-
ditions, cell-associated cytokines in peripheral blood
mononuclear cells (PBMCs) are decreased, and the ca-
pacity of these cells to produce tumor necrosis factor-␣
(TNF-␣) and interleukin-1␤ (IL-1␤) in vitro in response
to LPS is significantly reduced [7, 10]. Hyporespon-
siveness is not only present in monocytes, but it occurs
in whole blood [11] and is associated with increased
plasma levels of IL-10 and prostaglandin E2, which are
known to inhibit the production of proinflammatory cy-
tokines [10]. Terms such as monocyte deactivation, im-
munoparalysis, or more simply cell hyporesponsiveness
have been introduced in order to illustrate the incapacity
of cells to respond to LPS stimuli due to overproduction
 Ronco et al: Sorbent application in sepsis
Fig. 1. Schematic representation of the clinical presentation and biologic counterparts occurring in the proinflammatory and anti-inflammatory
phases of sepsis.
of anti-inflammatory cytokines (Fig. 1) [7, 11–13]. All
therapies specifically devoted to sepsis have so far ob-
tained frustrating results [6]. Liano et al prospectively
observed that ARF in the intensive care unit occurs
predominantly as part of a multiple organ failure, while
isolated ARF is the usual presentation in non-intensive
care unit patients [14]. Their study implied a higher mor-
tality in critically patients who developed ARF, support-
ing the concept of a direct effect of the ARF itself on
the mortality. Nevertheless, not only should therapeutic
strategies be applied to treat ARF itself, but any avail-
able technique should be involved in the possible preven-
tion of such derangement. This global approach probably
would help in modifying the course of the septic syn-
drome, and it may have an impact on mortality of the
critically ill patients.
CONTINUOUS RENAL
REPLACEMENT THERAPIES
Critically ill patients with ARF are preferably treated
with continuous renal replacement therapies (CRRTs)
[15, 16]. Conventional CRRTs such as hemofiltration
(CVVH) or hemodiafiltration (CVVHDF) have allowed
treatment of patients who could not be treated otherwise,
because of the remarkable hemodynamic instability and
the severe catabolic status [17]. Because of slow and
S-149
gentle ultrafiltration, CRRTs make it possible to control
fluid and electrolyte balance maintaining the patient in
steady hemodynamic conditions. Continuous fluid re-
moval allows infusion of large amounts of fluids and
ensures adequate caloric intake. In spite of these definite
advantages, the role of CRRTs in the framework of
“blood purification” has been challenged, and it is some-
times considered of minimal or no clinical relevance
[18, 19]. Nevertheless, the use of CRRTs is expanding,
and new potential advantages have been proposed.
The use of high permeability membranes in CRRTs
allows the removal of measurable quantities of cytokines,
although plasma levels seem to not be affected [19].
The mechanism of cytokine removal from the circulation
seems to rely on filtration and membrane adsorption.
Transport of sepsis-associated mediators across highly
permeable membranes may be largely unpredictable be-
cause of variable effects of filtration rates, molecular
interactions, presence of electric charges, hydrophilic or
hydrophobic sites on the membrane, and finally binding
to plasma proteins and/or acute phase reactants, as well
as to cell receptors. The possible elimination of cytokine-
inducing compounds rather than the cytokines should
also be considered when dealing with highly permeable
membranes. Hoffmann et al reported that hemofiltration
may remove cardiotoxic compounds [20] and TNF-␣–,
S-150 Ronco et al: Sorbent application in sepsis

Fig. 2. Molecular weights of compounds of interest in sepsis-associated ARF in respect to the cutoff of high-permeability synthetic membranes.

Fig. 3. Evolution of sorbents in extracorpo-

real therapies for acute and chronic renal
failure.

but not IL-6– or IL-1␤–inducing compounds [21]. Figure
2 summarizes the list of different mediators that are
known to be associated with sepsis. As can be seen, a
large part of these mediators has molecular weights
above the cutoff of highly permeable synthetic mem-
branes used in CRRTs. Therefore, considering all of
the previously mentioned factors, it seems that highly
permeable membranes may have a limited capacity of
removal for molecules involved in the pathogenesis of
sepsis. Because of the limited capacity of filtration and
 Ronco et al: Sorbent application in sepsis S-151

Table 1. The use of sorbents in clinical practice

Sorbent Type Modality Applications

Nonselective Charcoal
coated Hemoperfusion Poisoning
uncoated Hemodiafiltration Chronic renal failure
Uncharged resins (Amberlite XAD-7)
Selective Hydrophobic resins Coupled plasma filtration adsorption (CPFA) Rabdomiolysis, hepatic failure
Powdered sorbent Hemodiadsorption and regenerative HIV; drug overdose; hepatic
push-pull pheresis failure
Microphere-based detoxification system Regenerative push-pull pheresis Poisnoning, LDL apheresis
Engineered matrices: Polymyxin-B polystyrene-derivative fibers;
macroporous cellulosic beads;
Engineered matrices: polyethyleneimine macroporous cellulosic beads

the possible early saturation of the sites for adsorption, it
is very unlikely to obtain significant variations in plasma
circulating levels of the substances involved in the sys-
temic inflammatory syndrome.
INNOVATIVE TECHNIQUES IN CRRTS
Conventional CRRTs have gained increased popular-
ity because of their ability to remove the excess fluid
and sodium in septic patients. The efficiency in terms of
urea and other toxin removal is not questioned today.
Limitations seem to become evident for protein-bound
compounds or for solutes in the molecular range beyond
the membrane cutoff. In order to enhance the removal of
septic-associated mediators, two approaches have been
taken under consideration. In their original proposal,
Grootendorst et al suggested that high-volume (100 L/
day) hemofiltration could remove myocardial depressant
factors and other sepsis mediators [22]. Some of these
mediators could be found in the ultrafiltrate extracted
from a pig model of septic shock. In a subsequent clinical
study, Bellomo et al confirmed this possibility, as evi-
denced by a significant reduction in norepinephrine ad-
ministration in animals with septic shock [23].
A different approach has been to use membranes with
larger pores and permeability characteristics superior to
those currently used in hemofiltration. Lee et al sug-
gested an increased survival in a porcine model of septic
shock and related it to a general effect of enhanced blood
purification [24]. Other authors have reported either in
animal [25] or human [26] studies improved survival
when using plasmapheresis. In vitro studies show that
plasma filtration allows the removal of higher amounts
of proinflammatory cytokines such as TNF-␣, IL-1␤, and
IL-8. Clearance and sieving coefficients of these cyto-
kines are significantly increased with “open” plasma fil-
tration membranes in comparison to high-flux hemofil-
tration membranes [27]. Based on these considerations,
it would appear of great interest to expand the use of
plasmapheresis in septic patient and to study its impact
on survival on a larger scale. However, plasmapheresis
is hardly considerable as a CRRT. Large amounts of
plasma substitutes and the evident cost implications may,
in fact, limit the application of these techniques in the
clinical stage. Furthermore, plasma filtration techniques
may lead to unwanted losses of plasma constituents that
cannot be adequately replaced by substitution fluids.
The combined requirements of a CRRT with high
sieving capacity and possible selective removal of sepsis-
associated mediators seem to find an answer in the appli-
cation of sorbents.
SORBENTS IN EXTRACORPOREAL THERAPIES
The use of sorbents in extracorporeal therapies has
been applied to the management of both acute and
chronic renal failure (Fig. 3 and Table 1) [28]. Classically,
hemoperfusion (HP) has been described as a technique
in which the sorbent was placed in direct contact with
blood in an extracorporeal circulation. More recently,
this technique and other techniques like apheresis have
witnessed intensive research. Sorbents may be of natural
origin such as charcoal (mineral or vegetal) or synthetic
(different resins with covalently bound groups reactive
with specific ligands). Sorbents have been applied in
different treatment modalities such as HP, HP coupled
with hemodialysis (HPHD), double-chamber hemodia-
filtration (PFDsorb), or coupled plasma filtration-adsorp-
tion (CPFA; Fig. 4). HP has the advantage of a much
simpler circuit, but it requires a very biocompatible sor-
bent because of the direct contact with blood and with
blood cells in particular. Charcoal has a high adsorbing
capacity, especially for low molecular weight waste prod-
ucts that accumulate during kidney or liver failure. Its
use in HP, however, requires a coating of the sorbent
surface to make it biocompatible. Coated charcoal, al-
though biocompatible, has a remarkably reduced adsorp-
tive capacity because of the cutoff of the coating material.
More recently, synthetic polymers have been introduced
with remarkable capacity for adsorption. The size selec-
tivity is only offered by the size of the pores on the
surface of the granular elements and not by the material
S-152 Ronco et al: Sorbent application in sepsis

Fig. 4. Possible modes of application of sorbents. (Top) The sorbent unit is placed in series before the hemodiafilter. Blood comes in direct
contact with the sorbent, and high biocompatibility is required. The system is defined HPHD. (Middle) The sorbent unit is placed online in the
ultrafiltrate produced from an hemofilter. The hemofilter is placed in series with the hemodiafilter. The system is used for online hemodiafiltration
in chronic patients, and it is defined as paired filtration dialysis with sorbent (PFDsorb). (Bottom) The sorbent unit is placed online in the plasma
filtrate produced from a plasma filter. The plasma filter is placed in series with the hemodiafilter. The system is used for critically ill patients with
septic shock, and it is defined as coupled plasma filtration adsorption (CPFA).

Table 2. Plasma treatment in sepsis: Criteria to assess safety. In another technique using uncoated sorbents [detoxi-
In vitro fication plasma filtration (DTPF); HemoCleanse, Inc.,
• Absorption of proteins (e.g., coagulation factors) West Lafayette, IN, USA], a hemodiabsorption mecha-
• Adsorption of antibiotics
• Activation of fluid phase systems (e.g., complement, coagulation,
nism is associated with a push-pull plasma filtration sys-
fibrinolysis, kinins) tem (a suspension of powdered sorbents surrounding 0.5
• Toxicology tests micros plasma filter membranes). Bidirectional plasma
• Endotoxin content
• Particulate release flow (at 80 to 100 mL/min) across the plasma filtration
• Pressure drop membrane provides direct contact between plasma pro-
• Biocompatibility tests (ISO 9000) teins and powdered sorbents, as well as clearance of
• Risk analysis of the circuit
In vivo cytokines (TNF-␣, IL-1␤, and IL-6) [29].
• Mortality in control animals There has been a widespread tendency to remove “bad
factors” rather than to attempt to bring about a restora-
tion of balance of physiological factors. Often, too much
emphasis has been placed on individual markers. We
itself. Another approach consists of the use of sorbents suggest that treatments should focus more carefully on
in “uncoated” form. These, however, cannot be placed a “balancing hypothesis” trying to restore a correct equi-
in direct contact with whole blood, and they are used librium between immunologic suppression and activa-
for the treatment online of the ultrafiltrate or the plasma- tion.
filtrate [28]. In these systems, plasma or plasma water is This is particularly true when the converging concepts
separated from whole blood, and after passing through of plasma filtration coupled with adsorption are to be
the sorbent, they are reinfused into the blood circuit applied to the very complex scenario of severe sepsis
reconstituting whole blood structure. evolving in septic shock.
 Ronco et al: Sorbent application in sepsis
Fig. 5. Adsorption capacity for tumor necrosis factor-␣ (TNF-␣) of two different sorbents: (A) uncoated charcoal, and (B) hydrophobic resin.
The experimental conditions were as described in Tetta et al [27].
CONTINUOUS PLASMA
FILTRATION ADSORPTION
Continuous plasma filtration adsorption (CPFA) is a
modality of blood purification in which plasma is sepa-
rated from whole blood and circulated in a sorbent car-
tridge. After the sorbent unit, plasma is returned to the
blood circuit, and the whole blood undergoes hemofil-
tration or hemodialysis.
Rationale
The rationale consists in the attempt to achieve ade-
quate removal of molecules that are not removed by
other hemofiltration or hemodialysis techniques. The ra-
tionale for exposing the plasma to the sorbent in a plasma
filtration system is to exclude the blood cells from the
contact with the sorbent and to reinfuse endogenous
plasma after nonselective simultaneous removal of dif-
ferent sepsis-associated mediators without the need of
donor plasma. The main issue is concerning the sparing
effect on endogenous plasma as compared with potential
unwanted losses of autologous plasma compounds.
The interesting rationale for such application has stim-
ulated a series of in vitro studies [27] and in animals,
specifically on a model of rabbit endotoxic shock [30].
These studies also aimed at assessing safety (Table 2).
Various types of sorbents have been tested for this appli-
cation (Fig. 5).
In vitro studies
In vitro studies demonstrated that removal rates for
different molecules may be very different according to
the structure and nature of the used sorbent [27]. More
S-153
importantly, when tested at different linear velocity of
the cross flow, the efficiency in removing cytokines is
maintained in a wide range of flow rates. Increased cross
flow velocity may reduce the efficiency of the sorbent.
In spite of that, the amount of cytokine removed in
one passage is far above the overall amount of cytokine
carried into the sorbent cartridge (calculated on the basis
of the highest levels detected in the plasma and the blood
flows utilized in CRRT) by the blood from septic patients
[27]. These studies also showed that for hydrophodic
resins of a given particle and pore size the binding of
cytokines (TNF-␣) occurs after prior adsorption of ␣2-
macroglobulin, the carrier of cytokines in plasma.
A major criticism may be raised concerning the re-
moval of beneficial substances or drugs by the mecha-
nism of adsorption. By using the experimental conditions
described in Tetta et al [27], we assessed the different
adsorptive properties of a hydrophobic resin for the most
commonly used antibiotics (Fig. 6). Except for Vancomy-
cin, where a modest removal can be observed, the levels
of other antibiotics such as Tobramycin or Amikacin
tend to remain stable over time.
Animal studies
Animal studies were performed in a rabbit model of
septic shock [30]. The model consisted of a single intrave-
nous injection of LPS. The dose was experimentally ad-
justed to determine a mortality of 80% of the control
animals at 72 hours. Coupled plasma filtration-adsorp-
tion resulted in a significant (P ⫽ 0.0041) survival (85%)
at 72 hours with respect to untreated control rabbits
injected with the same amount of LPS.
In the pathogenesis of gram-negative infections, the
S-154 Ronco et al: Sorbent application in sepsis
complex and dynamic host interaction involves the in-
flammatory cascade, complement activation, coagulation
cascade, and hemodynamic derangements. In the animal
model, improved survival was negatively correlated with
the severity score, which included plasma LPS concentra-
tion, bioactive TNF, as well as mean arterial pressure
(MAP), base excess (BE), and white blood cell count
(WBC). However, cumulative survival was not correlated
with the levels of circulating TNF. It must be emphasized
that the overall net effect on survival could be due to
the removal not only of the measured mediators, but
also to many other mediators not monitored in our study.
The possibility that simultaneous removal of different
Fig. 6. Adsorption studies of antibiotics. Five
hundred milliliters of plasma were added in
separate experiments with Tobramycin (Nebi-
cina, Eli Lilly), vancomycin (Vancocina, Eli
Lilly) and Amikacin (BB K8, Bristol-Myers
Squibb) in order to reach the therapeutic con-
centrations. Vues recorded before (䉬) and
after (䊏) the cartridge were determined for
tobramycin (A) and amikacin (C) using the
PFIA Abbott analyzer TDX (Abbott), and
for vancomycin (B) using the PFIA Abbott
analyzer Axsym. Concentration values (pre-,
post-, and percentage removal) for tobra-
mycin, vancomycin, and amikacin were as fol-
lows: 10,600 ␮g, 8823 ␮g, 17%; 9267 ␮g, 300
␮g, 97%; and 9560 ␮g, 8260 ␮g, 10%. Other
agents (data not shown) such as gentamin and
teichoplanin were tested in the same experi-
mental conditions: The removal rates were 56
and 93%, respectively.
mediators could be linked in a cause-effect relationship
with improved survival in our experimental model was
suggested. Furthermore, the study provided the rationale
for the application of this new method of blood purifica-
tion in septic patients undergoing CRRT [31, 32].
Clinical trial
A prospective randomized crossover trial aimed at
comparing clinical and biological effects of CPFA versus
CVVH in critically ill septic patients has recently been
concluded. Preliminary results were presented in ab-
stract form (abstract; Brendolan et al, J Am Soc Nephrol
9:A0655, 1998). The major findings can be summarized as
follows: restoration of cell responsiveness to exogenous
 Ronco et al: Sorbent application in sepsis
LPS after five hours of treatment in all patients; increased
systemic vascular resistances and significant reduction of
the dose of norepinephrine required to maintain a stable
hemodynamics in the patients (mean 30%). These data
suggest the possibility that CPFA, as opposed to CVVH,
may ensure an improved hemodynamic response in
highly unstable patients. Since CPFA may be modular
to conventional CVVH, the two modalities can be car-
ried out in series. The system may ensure a fluid and
salt balance together with enhanced blood purification
for various molecules.
Efficiency and adequacy of treatment, known mile-
stones in the extracorporeal treatment for chronic renal
failure, are now reconsidered in critical care nephrology.
The complex scenario of sepsis must not be underesti-
mated. Notwithstanding, 20 years or so after the first
descriptions, we all face a disease with an ever-increasing
incidence and unacceptably high mortality.
Innovative techniques address the importance of dedi-
cated extracorporeal systems for sepsis where ARF is
just one of the pathologic complications.
This wider approach to the concept of blood purifica-
tion opens new perspectives in a revisited strategy for
the application of extracorporeal treatments.
Reprint requests to Claudio Ronco, M.D., Renal Research Institute,
207 East 94 Street, Suite 303, New York, New York 10128, USA.
E-mail: cronco@rriny.com
REFERENCES
1. Beaman M, Turney JH, Rodger RSC, McGonigle RJS, Adu D,
Michael J: Changing pattern of acute renal failure. Q J Med
237:15–23, 1987
2. Schaefer JH, Jochimsen F, Keller F, Wegscheider K, Distler
A: Outcome prediction of acute renal failure in medical intensive
care. Intensive Care Med 17:19–24, 1991
3. Brivet FG, Kleinknecht DJ, Liorat P, Landais PJM: Acute renal
failure in intensive care units: Causes, outcome, and prognostic
factors of hospital mortality: A prospective, multicenter study. Crit
Care Med 24:192–198, 1996
4. Morrison DC, Ulevitch RY: The effects of bacterial endotoxin
on host mediator system. Am J Pathol 93:527–617, 1987
5. Glauser MP, Zanetti G, Baumgartner JD, Cohen J: Septic
shock: Pathogenesis. Lancet 338:732–736, 1991
6. Zeni F, Freeman B, Nathanson C: Anti-inflammatory therapies
to treat sepsis and septic shock: A reassessment. Crit Care Med
25:1095–1100, 1997
7. Munoz C, Carlet J, Fitting C, Misset B, Bleriot JP, Cavaillon
JM: Dysregulation of in vitro cytokine production by monocytes
during sepsis. J Clin Invest 88:1747–1754, 1991
8. Randow F, Syrbe U, Meisel C, Krausch D, Zuckermann H,
Platzer C, Volk HD: Mechanism of endotoxin desensitization:
Involvement of interleukin-10 and transforming growth factor ␤.
J Exp Med 181:1887–1892, 1995
9. Gutierrez-Ramos JC, Bluethmann H: Molecules and mecha-
nisms operating in septic shock: Lessons from knockout mice.
Immunol Today 18:329–334, 1997
10. Astiz M, Saha D, Lustrader D, Lin R, Rackow E: Monocyte
response to bacterial toxins, expression of cell surface receptors,
and release of anti-inflammatory cytokines during sepsis. J Lab
Clin Med 128:594–600, 1996
11. Lonnemann G, Bechstein M, Linnenweber S, Burg M, Koch
KM: Tumor necrosis factor-␣ during continuous high-flux hemodi-
S-155
alysis in sepsis with acute renal failure. Kidney Int 56(Suppl 72):
S84–S87, 1999
12. Volk HD, Reinke P, Krausch D, Zuckermann H, Asadullah
K, Mueller JM, Doecke WD, Kox WJ: Monocyte deactivation:
Rationale for a new therapeutic strategy in sepsis. Intensive Care
Med 22:(Suppl 4):S474–S481, 1996
13. Brandtzaeg P, Osnes L, Ovstebo R, Joo GB, Westvik A-B,
Kierulf P: Net inflammatory capacity of human septic shock
plasma evaluated by a monocyte-based target cell assay: Identifica-
tion of interleukin-10 as a major functional deactivator of human
monocytes. J Exp Med 184:51–60, 1996
14. Liano F, Junco E, Pascual J, Madero R, Verde E, the Madrid
Acute Renal Failure Study Group: The spectrum of acute renal
failure in the intensive care unit compared with that seen in other
settings. Kidney Int 53(Suppl 66):S16–S24, 1998
15. Bellomo R, Farmer M, Wright C, Parkin G, Boyce N: Treatment
of sepsis-associated severe acute renal failure with continuous hem-
odiafiltration: Clinical experience and comparison with conven-
tional dialysis. Blood Purif 13:246–254, 1995
16. Silvester W, Bellomo R, Ronco C: Continuous versus intermit-
tent renal replacement therapy in the critically ill, in Critical Care
Nephrology, edited by Ronco C, Bellomo R, Dordrecht, Kluwer
Academic Publishers, 1998, pp 1225–1238
17. Canaud B, Mion C: Extracorporeal treatment of acute renal fail-
ure: Methods, indications, quantified and personalized therapeutic
approach. Adv Nephrol 24:271–281, 1995
18. Lameire N, Van Biesen W, Vanholder R: Dialysing the patient
with acute renal failure in the ICU: The emperor’s clothes? Nephrol
Dial Transplant 14:2570–2573, 1999
19. de Vriese A, Colardyn A, Philippé JI, Vanholder RC, De Sutter
JH, Lameire N: Cytokine removal during continuous hemofiltra-
tion in septic patients. J Am Soc Nephrol 10:846–853, 1999
20. Hoffmann JN, Verdan K, Hartl WH, Jochum M, Faist E, In-
thorn D: Hemofiltrate from patients with severe sepsis and de-
pressed left ventricular contractility contains cardiotoxic com-
pounds. Shock 12:174–180, 1999
21. Hoffmann JN, Hartl WH, Deppisch R, Faist E, Jochum M, In-
thorn D: Hemofiltration in human sepsis: Evidence for elimination
of immunomodulatory substances. Kidney Int 48:1563–1570, 1995
22. Grootendorst AF, Van Bommel EF, Van Leengoed LA, Nabu-
urs M, Bouman CS, Groeneveld AB: High, hemofiltration im-
proves right ventricular function in endotoxin-induced shock in
the pig. Intensive Care Med 18:235–240, 1992
23. Bellomo R, Baldwin I, Cole L, Ronco C: Preliminary experience
with high-volume hemofiltration in human septic shock. Kidney
Int 53(Suppl 66):S182–S185, 1998
24. Lee Pa, Matson Jr, Pryor RW, Hinshaw LB: Continuous arterio-
venous hemofiltration therapy for Staphylococcus aureus-induced
septicemia in immature swine. Crit Care Med 21:914–924, 1993
25. Lonergan JM, Orlowski JP, Sato T, Mchugh MJ, Zborowski
M: Extracorporeal endotoxin removal in a canine model of septic
shock. ASAIO J 40:M654–M657, 1994
26. Berlot G, Tomasini A, Silvestri L, Gullo A: Plasmapheresis in
the critically ill patient. Kidney Int 53(Suppl 66):S178–S181, 1998
27. Tetta C, Cavaillon JM, Schulze M, Ronco C, Ghezzi PM,
Camussi G, Serra AM, Curti F, Lonnemann G: Removal of
cytokines and activated complement components in an experimen-
tal model of continuous plasmafiltration coupled with sorbent ad-
sorption. Nephrol Dial Transplant 13:1458–1464, 1998
28. La Greca G, Brendolan A, Ghezzi PM, De Smet R, Tetta C,
Gervasio R, Ronco C: The concept of sorbents in hemodialysis.
Int J Artif Organs 21:303–308, 1998
29. Levy H, Ash SR, Knab W, Steczko J, Carr DJ, Blake DE:
Systemic inflammatory response syndrome treatment by powdered
sorbent pheresis: The biologic-detoxification plasma filtration sys-
tem. ASAIO J 44:M659–M665, 1998
30. Tetta C, Gianotti L, Cavaillon JM, Wratten ML, Fini M, Braga
M, Bisagni P, Giavaresi GL, Bolzani R, Giardino R: Coupled
plasmafiltration-adsorption in a rabbit model of endotoxic shock.
Crit Care Med (in press)
31. Parillo JE: Pathogenetic mechanisms of septic shock. N Engl J
Med 328:1471–1477, 1993
32. Silvester W: Mediator removal with CRRT: Complement and
cytokines. (review) Am J Kidney Dis 30(5 Suppl 4):S38–S43, 1997